diff --git a/.github/workflows/benchmark_asterinas.yml b/.github/workflows/benchmark_asterinas.yml
index aca2555b..225ed281 100644
--- a/.github/workflows/benchmark_asterinas.yml
+++ b/.github/workflows/benchmark_asterinas.yml
@@ -180,24 +180,29 @@ jobs:
           pattern: results_*
           path: ./results
           merge-multiple: true
+      
+      - name: Set up the environment
+        run: |
+          sudo apt-get update && sudo apt-get install -y yq
 
       - name: Set up benchmark configuration
         run: |
           BENCHMARK_DIR=$(echo ${{ matrix.benchmark }} | sed 's/-/\//g')
           BENCHMARK_SUITE=$(echo $BENCHMARK_DIR | awk -F'/' '{print $1}')
           BENCHMARK_NAME=$(echo "$BENCHMARK_DIR" | sed -E 's|^[^/]+/||; s|/bench_results||g; s|/|_|g')
-          BENCH_RESULT="test/benchmark/${BENCHMARK_DIR}/bench_result.json"
-          [ -f "$BENCH_RESULT" ] || BENCH_RESULT="test/benchmark/${BENCHMARK_DIR}.json"
-          ALERT_THRESHOLD=$(jq -r '.alert.threshold // "130%"' "$BENCH_RESULT")
-          ALERT_TOOL=$(jq -r 'if .alert.bigger_is_better == true then "customBiggerIsBetter" else "customSmallerIsBetter" end' "$BENCH_RESULT")
-          TITLE=$(jq -r '.chart.title // "Undefined"' "$BENCH_RESULT")
-          DESCRIPTION=$(jq -r '.chart.description // "No description provided"' "$BENCH_RESULT")
+          BENCH_RESULT="test/benchmark/${BENCHMARK_DIR}/bench_result.yaml"
+          [ -f "$BENCH_RESULT" ] || BENCH_RESULT="test/benchmark/${BENCHMARK_DIR}.yaml"
+          ALERT_THRESHOLD=$(yq -r '.alert.threshold // "130%"' "$BENCH_RESULT")
+          ALERT_TOOL=$(yq -r 'if (.alert.bigger_is_better == true) then "customBiggerIsBetter" else "customSmallerIsBetter" end' "$BENCH_RESULT")
+          TITLE=$(yq -r '.chart.title // "Undefined"' "$BENCH_RESULT")
+          DESCRIPTION=$(yq -r '.chart.description // "No description provided"' "$BENCH_RESULT")
           echo "BENCHMARK_SUITE=$BENCHMARK_SUITE" >> $GITHUB_ENV
           echo "BENCHMARK_NAME=$BENCHMARK_NAME" >> $GITHUB_ENV
           echo "ALERT_THRESHOLD=$ALERT_THRESHOLD" >> $GITHUB_ENV
           echo "ALERT_TOOL=$ALERT_TOOL" >> $GITHUB_ENV
           echo "TITLE=$TITLE" >> $GITHUB_ENV
           echo "DESCRIPTION=$DESCRIPTION" >> $GITHUB_ENV
+          yq . test/benchmark/$BENCHMARK_SUITE/summary.yaml > test/benchmark/$BENCHMARK_SUITE/summary.json
 
       - name: Store benchmark results
         uses: asterinas/github-action-benchmark@v4
diff --git a/test/benchmark/README.md b/test/benchmark/README.md
index bf8eb96f..49341f2a 100644
--- a/test/benchmark/README.md
+++ b/test/benchmark/README.md
@@ -67,11 +67,11 @@ For jobs that produce a single result, the directory is structured as follows:
 ```plaintext
 <bench_suite>/
 └── <single_result_job>/
-    ├── bench_result.json
+    ├── bench_result.yaml
     └── run.sh
 ```
 
-The `bench_result.json` file contains metadata about the result, including the title, measurement unit, and whether higher or lower values indicate better performance.
+The `bench_result.yaml` file contains metadata about the result, including the title, measurement unit, and whether higher or lower values indicate better performance.
 
 #### Multi-Result Jobs
 
@@ -81,12 +81,12 @@ For jobs producing multiple results, the directory includes a `bench_results/` f
 <bench_suite>/
 └── <multi_result_job>/
     ├── bench_results/
-    │   ├── <job_a>.json
-    │   └── <job_b>.json
+    │   ├── <job_a>.yaml
+    │   └── <job_b>.yaml
     └── run.sh
 ```
 
-Each JSON file in the `bench_results/` directory describes a specific result's metadata.
+Each YAML file in the `bench_results/` directory describes a specific result's metadata.
 
 ## Adding New Benchmark Jobs
 
@@ -102,7 +102,7 @@ Each benchmark job should be added under the corresponding suite in the `test/be
 <bench_suite>/
 └── <job>/
     ├── host.sh # Only for host-guest jobs
-    ├── bench_result.json  # or bench_results/ directory for multiple results jobs
+    ├── bench_result.yaml  # or bench_results/ directory for multiple results jobs
     └── run.sh
 ```
 
@@ -134,53 +134,46 @@ iperf3 -c $GUEST_SERVER_IP_ADDRESS -f m
 
 #### Configuration Files
 
-The configuration files provide metadata about the benchmark jobs and results, such as the regression alerts, chart details, and result extraction patterns. Typically, these files are in JSON format. For single-result jobs, a `bench_result.json` file is used, while multi-result jobs have individual JSON files under `bench_results/` for each result. Some fields in these files are necessary while some are optional, depending on the benchmark's requirements. For more information, see the [`bench_result.json` format](#the-bench_resultjson-format) section.
+The configuration files provide metadata about the benchmark jobs and results, such as the regression alerts, chart details, and result extraction patterns. Typically, these files are in YAML format. For single-result jobs, a `bench_result.yaml` file is used, while multi-result jobs have individual YAML files under `bench_results/` for each result. Some fields in these files are necessary while some are optional, depending on the benchmark's requirements. For more information, see the [`bench_result.yaml` format](#the-bench_resultyaml-format) section.
 
 Below are the contents of these files for the sample benchmark:
 
-```jsonc
-// fio/ext2_no_iommu_seq_write_bw/bench_result.json
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "bw=",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Ext2] The bandwidth of sequential writes (IOMMU disabled on Asterinas)",
-        "description": "fio -filename=/ext2/fio-test -size=1G -bs=1M -direct=1",
-        "unit": "MB/s",
-        "legend": "Average file write bandwidth on {system}"
-    },
-    "runtime_config": {
-        "aster_scheme": "null"
-    }
-}  
+```yaml
+# fio/ext2_no_iommu_seq_write_bw/bench_result.yaml
+alert:
+  threshold: "125%"
+  bigger_is_better: true
+
+result_extraction:
+  search_pattern: "bw="
+  result_index: 2
+
+chart:
+  title: "[Ext2] The bandwidth of sequential writes (IOMMU disabled on Asterinas)"
+  description: "fio -filename=/ext2/fio-test -size=1G -bs=1M -direct=1"
+  unit: "MB/s"
+  legend: "Average file write bandwidth on {system}"
+
+runtime_config:
+  aster_scheme: "null"
 ```
 
-```jsonc
-// sqlite/ext2_benchmarks/bench_results/ext2_deletes_between.json
-{
-    "result_extraction": {
-        "search_pattern": "[0-9]+ DELETEs, numeric BETWEEN, indexed....",
-        "result_index": 8
-    },
-    "chart": {
-    ...
-}
+```yaml
+# sqlite/ext2_benchmarks/bench_results/ext2_deletes_between.yaml
+result_extraction:
+  search_pattern: "[0-9]+ DELETEs, numeric BETWEEN, indexed...."
+  result_index: 8
 
-// sqlite/ext2_benchmarks/bench_results/ext2_updates_between.json
-{
-    "result_extraction": {
-        "search_pattern": "[0-9]+ UPDATES, numeric BETWEEN, indexed....",
-        "result_index": 8
-    },
-    "chart": {
-    ...
-}
+chart:
+  title: "SQLite Ext2 Deletes Between"
+
+# sqlite/ext2_benchmarks/bench_results/ext2_updates_between.yaml
+result_extraction:
+  search_pattern: "[0-9]+ UPDATES, numeric BETWEEN, indexed...."
+  result_index: 8
+
+chart:
+...
 ```
 
 ### Step 3: Update Suite's `summary.json`
@@ -229,36 +222,28 @@ Secondly, we can validate modifications by running the CI pipeline on our own re
 
 Finally, if the new benchmark job runs successfully, we can commit the changes and create a pull request to merge the new benchmark into the main branch.
 
-## The `bench_result.json` Format
+## The `bench_result.yaml` Format
 
-The `bench_result.json` file configures how benchmark results are processed and displayed. Below is an example of the file to give you a big-picture understanding:
+The `bench_result.yaml` file configures how benchmark results are processed and displayed. Below is an example of the file to give you a big-picture understanding:
 
-```jsonc
-{
-    // Configurations for performance alerts.
-    "alert": {
-        "threshold": "130%", // Acceptable performance deviation (e.g., 130% = 30% higher).
-        "bigger_is_better": true // true: Higher values are better; false: Lower values are better.
-    },
-    // Settings for extracting benchmark results from raw outputs.
-    "result_extraction": {
-        "search_pattern": "sender", // Regex or string to locate results.
-        "nth_occurrence": 1, // Optional. Which matched occurrence to use (default to 1).
-        "result_index": 7 // Match index to use.
-    },
-    // Configurations for how the results are displayed in charts.
-    "chart": {
-        "title": "[Network] iperf3 sender performance using TCP", // Title of the chart.
-        "description": "iperf3 -s -B 10.0.2.15", // Context or command associated with the benchmark.
-        "unit": "Mbits/sec", // Measurement unit for the results.
-        "legend": "Average TCP Bandwidth over virtio-net between Host Linux and Guest {system}" // Chart legend with dynamic placeholder {system} supported.
-    },
-    // Optional runtime configurations for the benchmark.
-    "runtime_config": {
-        "aster_scheme": "null", // Corresponds to Makefile parameters, IOMMU is enabled by default (SCHEME=iommu).
-        "smp": 1 // Number of SMP CPUs (default to 1).
-    }
-}
+```yaml
+alert:                        # Alert configuration for performance regression
+  threshold: "130%"           # Acceptable performance deviation (e.g., 130% = 30% higher)
+  bigger_is_better: true      # true: Higher values are better; false: Lower values are better
+
+result_extraction:            # Result extraction configuration
+  search_pattern: "sender"    # Regex or string to locate results
+  nth_occurrence: 1           # Optional. Which matched occurrence to use (default to 1).
+  result_index: 7             # Match index to use
+
+chart:                        # Chart configuration
+  title: "[Network] iperf3 sender performance using TCP"           # Title of the chart
+  description: "iperf3 -s -B 10.0.2.15"                            # Context or command associated with the benchmark
+  unit: "Mbits/sec"                                                # Measurement unit for the results
+  legend: "Average TCP Bandwidth over virtio-net between Host Linux and Guest {system}" # Chart legend with dynamic placeholder {system} supported
+
+runtime_config:              # Runtime configuration
+  aster_scheme: "null"       # Corresponds to Makefile parameters, IOMMU is enabled by default (SCHEME=iommu)
 ```
 
 By adhering to this format, we ensure clarity and consistency in benchmarking workflows and reporting systems.
diff --git a/test/benchmark/bench_linux_and_aster.sh b/test/benchmark/bench_linux_and_aster.sh
index 883ce407..afd3790e 100755
--- a/test/benchmark/bench_linux_and_aster.sh
+++ b/test/benchmark/bench_linux_and_aster.sh
@@ -6,7 +6,7 @@ set -e
 set -o pipefail
 
 # Ensure all dependencies are installed
-command -v jq >/dev/null 2>&1 || { echo >&2 "jq is not installed. Aborting."; exit 1; }
+command -v yq >/dev/null 2>&1 || command -v yq >/dev/null 2>&1 || { echo >&2 "tools are not installed. Aborting."; exit 1; }
 
 # Set up paths
 BENCHMARK_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")" &>/dev/null && pwd)"
@@ -32,10 +32,11 @@ parse_raw_results() {
     fi
 
     # Write the results into the template
-    jq --arg linux_result "${linux_result}" --arg aster_result "${aster_result}" \
+    yq --arg linux_result "${linux_result}" --arg aster_result "${aster_result}" \
         '(.[] | select(.extra == "linux_result") | .value) |= $linux_result |
          (.[] | select(.extra == "aster_result") | .value) |= $aster_result' \
         "${RESULT_TEMPLATE}" > "${result_file}"
+    echo "Results written to ${result_file}"
 }
 
 # Generate a new result template based on unit and legend
@@ -48,7 +49,7 @@ generate_template() {
     local asterinas_legend=${legend//"{system}"/"Asterinas"}
 
     # Generate the result template JSON
-    jq -n --arg linux "$linux_legend" --arg aster "$asterinas_legend" --arg unit "$unit" '[
+    yq -n --arg linux "$linux_legend" --arg aster "$asterinas_legend" --arg unit "$unit" '[
         { "name": $linux, "unit": $unit, "value": 0, "extra": "linux_result" },
         { "name": $aster, "unit": $unit, "value": 0, "extra": "aster_result" }
     ]' > "${RESULT_TEMPLATE}"
@@ -66,7 +67,8 @@ extract_result_file() {
         local second_part=$(dirname "$bench_result" | awk -F"/benchmark/$first_dir/" '{print $2}' | cut -d'/' -f1)
         echo "result_${first_dir}-${second_part}.json"
     else
-        echo "result_${relative_path//\//-}"
+        local result_file="result_${relative_path//\//-}"
+        echo "${result_file/.yaml/.json}"
     fi
 }
 
@@ -138,11 +140,11 @@ run_benchmark() {
 parse_results() {
     local bench_result="$1"
 
-    local search_pattern=$(jq -r '.result_extraction.search_pattern // empty' "$bench_result")
-    local nth_occurrence=$(jq -r '.result_extraction.nth_occurrence // 1' "$bench_result")
-    local result_index=$(jq -r '.result_extraction.result_index // empty' "$bench_result")
-    local unit=$(jq -r '.chart.unit // empty' "$bench_result")
-    local legend=$(jq -r '.chart.legend // {system}' "$bench_result")
+    local search_pattern=$(yq -r '.result_extraction.search_pattern // empty' "$bench_result")
+    local nth_occurrence=$(yq -r '.result_extraction.nth_occurrence // 1' "$bench_result")
+    local result_index=$(yq -r '.result_extraction.result_index // empty' "$bench_result")
+    local unit=$(yq -r '.chart.unit // empty' "$bench_result")
+    local legend=$(yq -r '.chart.legend // {system}' "$bench_result")
 
     generate_template "$unit" "$legend"
     parse_raw_results "$search_pattern" "$nth_occurrence" "$result_index" "$(extract_result_file "$bench_result")"
@@ -167,22 +169,22 @@ main() {
     local run_mode="guest_only"
     [[ -f "${BENCHMARK_ROOT}/${benchmark}/host.sh" ]] && run_mode="host_guest"
 
-    local bench_result="${BENCHMARK_ROOT}/${benchmark}/bench_result.json"
+    local bench_result="${BENCHMARK_ROOT}/${benchmark}/bench_result.yaml"
     local aster_scheme
     if [[ -f "$bench_result" ]]; then
-        aster_scheme=$(jq -r '.runtime_config.aster_scheme // ""' "$bench_result")
+        aster_scheme=$(yq -r '.runtime_config.aster_scheme // ""' "$bench_result")
     else
         for job in "${BENCHMARK_ROOT}/${benchmark}"/bench_results/*; do
-            [[ -f "$job" ]] && aster_scheme=$(jq -r '.runtime_config.aster_scheme // ""' "$job") && break
+            [[ -f "$job" ]] && aster_scheme=$(yq -r '.runtime_config.aster_scheme // ""' "$job") && break
         done
     fi
 
     local smp
     if [[ -f "$bench_result" ]]; then
-        smp=$(jq -r '.runtime_config.smp // 1' "$bench_result")
+        smp=$(yq -r '.runtime_config.smp // 1' "$bench_result")
     else
         for job in "${BENCHMARK_ROOT}/${benchmark}"/bench_results/*; do
-            [[ -f "$job" ]] && smp=$(jq -r '.runtime_config.smp // 1' "$job") && break
+            [[ -f "$job" ]] && smp=$(yq -r '.runtime_config.smp // 1' "$job") && break
         done
     fi
 
diff --git a/test/benchmark/fio/ext2_seq_read_bw/bench_result.json b/test/benchmark/fio/ext2_seq_read_bw/bench_result.json
deleted file mode 100644
index c82e952d..00000000
--- a/test/benchmark/fio/ext2_seq_read_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "bw=[0-9]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Ext2] The bandwidth of sequential reads",
-        "description": "fio -filename=/ext2/fio-test -size=1G -bs=1M -direct=1",
-        "unit": "MB/s",
-        "legend": "Average file read bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/fio/ext2_seq_read_bw/bench_result.yaml b/test/benchmark/fio/ext2_seq_read_bw/bench_result.yaml
new file mode 100644
index 00000000..a115b4ef
--- /dev/null
+++ b/test/benchmark/fio/ext2_seq_read_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: fio -filename=/ext2/fio-test -size=1G -bs=1M -direct=1
+  legend: Average file read bandwidth on {system}
+  title: '[Ext2] The bandwidth of sequential reads'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: bw=[0-9]+
diff --git a/test/benchmark/fio/ext2_seq_read_bw_no_iommu/bench_result.json b/test/benchmark/fio/ext2_seq_read_bw_no_iommu/bench_result.json
deleted file mode 100644
index 6404487c..00000000
--- a/test/benchmark/fio/ext2_seq_read_bw_no_iommu/bench_result.json
+++ /dev/null
@@ -1,19 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "bw=[0-9]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Ext2] The bandwidth of sequential reads (IOMMU disabled on Asterinas)",
-        "description": "fio -filename=/ext2/fio-test -size=1G -bs=1M -direct=1",
-        "unit": "MB/s",
-        "legend": "Average file read bandwidth on {system}"
-    },
-    "runtime_config": {
-        "aster_scheme": "null"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/fio/ext2_seq_read_bw_no_iommu/bench_result.yaml b/test/benchmark/fio/ext2_seq_read_bw_no_iommu/bench_result.yaml
new file mode 100644
index 00000000..ff2389d0
--- /dev/null
+++ b/test/benchmark/fio/ext2_seq_read_bw_no_iommu/bench_result.yaml
@@ -0,0 +1,13 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: fio -filename=/ext2/fio-test -size=1G -bs=1M -direct=1
+  legend: Average file read bandwidth on {system}
+  title: '[Ext2] The bandwidth of sequential reads (IOMMU disabled on Asterinas)'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: bw=[0-9]+
+runtime_config:
+  aster_scheme: 'null'
diff --git a/test/benchmark/fio/ext2_seq_write_bw/bench_result.json b/test/benchmark/fio/ext2_seq_write_bw/bench_result.json
deleted file mode 100644
index 6d6fcef3..00000000
--- a/test/benchmark/fio/ext2_seq_write_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "bw=[0-9]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Ext2] The bandwidth of sequential writes",
-        "description": "fio -filename=/ext2/fio-test -size=1G -bs=1M -direct=1",
-        "unit": "MB/s",
-        "legend": "Average file write bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/fio/ext2_seq_write_bw/bench_result.yaml b/test/benchmark/fio/ext2_seq_write_bw/bench_result.yaml
new file mode 100644
index 00000000..788d7d29
--- /dev/null
+++ b/test/benchmark/fio/ext2_seq_write_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: fio -filename=/ext2/fio-test -size=1G -bs=1M -direct=1
+  legend: Average file write bandwidth on {system}
+  title: '[Ext2] The bandwidth of sequential writes'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: bw=[0-9]+
diff --git a/test/benchmark/fio/ext2_seq_write_bw_no_iommu/bench_result.json b/test/benchmark/fio/ext2_seq_write_bw_no_iommu/bench_result.json
deleted file mode 100644
index efefea70..00000000
--- a/test/benchmark/fio/ext2_seq_write_bw_no_iommu/bench_result.json
+++ /dev/null
@@ -1,19 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "bw=[0-9]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Ext2] The bandwidth of sequential writes (IOMMU disabled on Asterinas)",
-        "description": "fio -filename=/ext2/fio-test -size=1G -bs=1M -direct=1",
-        "unit": "MB/s",
-        "legend": "Average file write bandwidth on {system}"
-    },
-    "runtime_config": {
-        "aster_scheme": "null"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/fio/ext2_seq_write_bw_no_iommu/bench_result.yaml b/test/benchmark/fio/ext2_seq_write_bw_no_iommu/bench_result.yaml
new file mode 100644
index 00000000..4ee456b8
--- /dev/null
+++ b/test/benchmark/fio/ext2_seq_write_bw_no_iommu/bench_result.yaml
@@ -0,0 +1,13 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: fio -filename=/ext2/fio-test -size=1G -bs=1M -direct=1
+  legend: Average file write bandwidth on {system}
+  title: '[Ext2] The bandwidth of sequential writes (IOMMU disabled on Asterinas)'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: bw=[0-9]+
+runtime_config:
+  aster_scheme: 'null'
diff --git a/test/benchmark/fio/summary.json b/test/benchmark/fio/summary.json
deleted file mode 100644
index 6b184aee..00000000
--- a/test/benchmark/fio/summary.json
+++ /dev/null
@@ -1,8 +0,0 @@
-{
-    "benchmarks": [
-        "ext2_seq_write_bw",
-        "ext2_seq_read_bw",
-        "ext2_seq_write_bw_no_iommu",
-        "ext2_seq_read_bw_no_iommu"
-    ]
-}
\ No newline at end of file
diff --git a/test/benchmark/fio/summary.yaml b/test/benchmark/fio/summary.yaml
new file mode 100644
index 00000000..b945c5b8
--- /dev/null
+++ b/test/benchmark/fio/summary.yaml
@@ -0,0 +1,5 @@
+benchmarks:
+  - ext2_seq_write_bw
+  - ext2_seq_read_bw
+  - ext2_seq_write_bw_no_iommu
+  - ext2_seq_read_bw_no_iommu
diff --git a/test/benchmark/hackbench/group8_smp1/bench_result.json b/test/benchmark/hackbench/group8_smp1/bench_result.json
deleted file mode 100644
index 0d9a9135..00000000
--- a/test/benchmark/hackbench/group8_smp1/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Time:",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Scheduler] Turnaround time of 8 groups using 1 CPU",
-        "description": "hackbench -g 8 -l 1000 -p -T",
-        "unit": "sec",
-        "legend": "Turnaround time of hackbench on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/hackbench/group8_smp1/bench_result.yaml b/test/benchmark/hackbench/group8_smp1/bench_result.yaml
new file mode 100644
index 00000000..95b52b37
--- /dev/null
+++ b/test/benchmark/hackbench/group8_smp1/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: hackbench -g 8 -l 1000 -p -T
+  legend: Turnaround time of hackbench on {system}
+  title: '[Scheduler] Turnaround time of 8 groups using 1 CPU'
+  unit: sec
+result_extraction:
+  result_index: 2
+  search_pattern: 'Time:'
diff --git a/test/benchmark/hackbench/group8_smp16/bench_result.json b/test/benchmark/hackbench/group8_smp16/bench_result.json
deleted file mode 100644
index 6898995f..00000000
--- a/test/benchmark/hackbench/group8_smp16/bench_result.json
+++ /dev/null
@@ -1,19 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Time:",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Scheduler] Turnaround time of 8 groups using 16 CPU",
-        "description": "hackbench -g 8 -l 1000 -p -T",
-        "unit": "sec",
-        "legend": "Turnaround time of hackbench on {system}"
-    },
-    "runtime_config": {
-        "smp": 16
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/hackbench/group8_smp16/bench_result.yaml b/test/benchmark/hackbench/group8_smp16/bench_result.yaml
new file mode 100644
index 00000000..eced288c
--- /dev/null
+++ b/test/benchmark/hackbench/group8_smp16/bench_result.yaml
@@ -0,0 +1,13 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: hackbench -g 8 -l 1000 -p -T
+  legend: Turnaround time of hackbench on {system}
+  title: '[Scheduler] Turnaround time of 8 groups using 16 CPU'
+  unit: sec
+result_extraction:
+  result_index: 2
+  search_pattern: 'Time:'
+runtime_config:
+  smp: 16
diff --git a/test/benchmark/hackbench/group8_smp8/bench_result.json b/test/benchmark/hackbench/group8_smp8/bench_result.json
deleted file mode 100644
index 1801107d..00000000
--- a/test/benchmark/hackbench/group8_smp8/bench_result.json
+++ /dev/null
@@ -1,19 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Time:",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Scheduler] Turnaround time of 8 groups using 8 CPU",
-        "description": "hackbench -g 8 -l 1000 -p -T",
-        "unit": "sec",
-        "legend": "Turnaround time of hackbench on {system}"
-    },
-    "runtime_config": {
-        "smp": 8
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/hackbench/group8_smp8/bench_result.yaml b/test/benchmark/hackbench/group8_smp8/bench_result.yaml
new file mode 100644
index 00000000..143d7cc9
--- /dev/null
+++ b/test/benchmark/hackbench/group8_smp8/bench_result.yaml
@@ -0,0 +1,13 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: hackbench -g 8 -l 1000 -p -T
+  legend: Turnaround time of hackbench on {system}
+  title: '[Scheduler] Turnaround time of 8 groups using 8 CPU'
+  unit: sec
+result_extraction:
+  result_index: 2
+  search_pattern: 'Time:'
+runtime_config:
+  smp: 8
diff --git a/test/benchmark/hackbench/summary.json b/test/benchmark/hackbench/summary.json
deleted file mode 100644
index 335faa68..00000000
--- a/test/benchmark/hackbench/summary.json
+++ /dev/null
@@ -1,4 +0,0 @@
-{
-    "benchmarks": [
-    ]
-}
\ No newline at end of file
diff --git a/test/benchmark/hackbench/summary.yaml b/test/benchmark/hackbench/summary.yaml
new file mode 100644
index 00000000..6953ccbd
--- /dev/null
+++ b/test/benchmark/hackbench/summary.yaml
@@ -0,0 +1 @@
+benchmarks: []
diff --git a/test/benchmark/iperf3/summary.json b/test/benchmark/iperf3/summary.json
deleted file mode 100644
index c0a26139..00000000
--- a/test/benchmark/iperf3/summary.json
+++ /dev/null
@@ -1,5 +0,0 @@
-{
-    "benchmarks": [
-        "tcp_virtio_bw"
-    ]
-}
\ No newline at end of file
diff --git a/test/benchmark/iperf3/summary.yaml b/test/benchmark/iperf3/summary.yaml
new file mode 100644
index 00000000..020abf12
--- /dev/null
+++ b/test/benchmark/iperf3/summary.yaml
@@ -0,0 +1,2 @@
+benchmarks:
+  - tcp_virtio_bw
diff --git a/test/benchmark/iperf3/tcp_virtio_bw/bench_result.json b/test/benchmark/iperf3/tcp_virtio_bw/bench_result.json
deleted file mode 100644
index 94e10061..00000000
--- a/test/benchmark/iperf3/tcp_virtio_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "sender",
-        "result_index": 7
-    },
-    "chart": {
-        "title": "[Network] iperf3 sender performance using TCP (virtio-net)",
-        "description": "iperf3 -s -B 10.0.2.15",
-        "unit": "Mbits/sec",
-        "legend": "Average TCP Bandwidth over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/iperf3/tcp_virtio_bw/bench_result.yaml b/test/benchmark/iperf3/tcp_virtio_bw/bench_result.yaml
new file mode 100644
index 00000000..1833e5da
--- /dev/null
+++ b/test/benchmark/iperf3/tcp_virtio_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 130%
+chart:
+  description: iperf3 -s -B 10.0.2.15
+  legend: Average TCP Bandwidth over virtio-net between Host Linux and Guest {system}
+  title: '[Network] iperf3 sender performance using TCP (virtio-net)'
+  unit: Mbits/sec
+result_extraction:
+  result_index: 7
+  search_pattern: sender
diff --git a/test/benchmark/lmbench/ext2_copy_files_bw/bench_result.json b/test/benchmark/lmbench/ext2_copy_files_bw/bench_result.json
deleted file mode 100644
index 3f95219d..00000000
--- a/test/benchmark/lmbench/ext2_copy_files_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "lmdd result: +[0-9.]+",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ext2] The bandwidth of copying data between files",
-        "description": "lmdd",
-        "unit": "MB/s",
-        "legend": "Average file copy bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/ext2_copy_files_bw/bench_result.yaml b/test/benchmark/lmbench/ext2_copy_files_bw/bench_result.yaml
new file mode 100644
index 00000000..f94e7daf
--- /dev/null
+++ b/test/benchmark/lmbench/ext2_copy_files_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: lmdd
+  legend: Average file copy bandwidth on {system}
+  title: '[Ext2] The bandwidth of copying data between files'
+  unit: MB/s
+result_extraction:
+  result_index: 8
+  search_pattern: 'lmdd result: +[0-9.]+'
diff --git a/test/benchmark/lmbench/ext2_create_delete_files_0k_ops/bench_result.json b/test/benchmark/lmbench/ext2_create_delete_files_0k_ops/bench_result.json
deleted file mode 100644
index 3dec1e35..00000000
--- a/test/benchmark/lmbench/ext2_create_delete_files_0k_ops/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "^0k\t[0-9]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Ext2] The throughput of creating/deleting small files (0KB)",
-        "description": "lat_fs -s 0k /ext2",
-        "unit": "number",
-        "legend": "Number of created/deleted files on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/ext2_create_delete_files_0k_ops/bench_result.yaml b/test/benchmark/lmbench/ext2_create_delete_files_0k_ops/bench_result.yaml
new file mode 100644
index 00000000..6a2bd77f
--- /dev/null
+++ b/test/benchmark/lmbench/ext2_create_delete_files_0k_ops/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: lat_fs -s 0k /ext2
+  legend: Number of created/deleted files on {system}
+  title: '[Ext2] The throughput of creating/deleting small files (0KB)'
+  unit: number
+result_extraction:
+  result_index: 2
+  search_pattern: ^0k\t[0-9]+
diff --git a/test/benchmark/lmbench/ext2_create_delete_files_10k_ops/bench_result.json b/test/benchmark/lmbench/ext2_create_delete_files_10k_ops/bench_result.json
deleted file mode 100644
index 131c580a..00000000
--- a/test/benchmark/lmbench/ext2_create_delete_files_10k_ops/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "^10k\t[0-9]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Ext2] The throughput of creating/deleting small files (10KB)",
-        "description": "lat_fs -s 10K /ext2",
-        "unit": "number",
-        "legend": "Number of created/deleted files on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/ext2_create_delete_files_10k_ops/bench_result.yaml b/test/benchmark/lmbench/ext2_create_delete_files_10k_ops/bench_result.yaml
new file mode 100644
index 00000000..151f2e6e
--- /dev/null
+++ b/test/benchmark/lmbench/ext2_create_delete_files_10k_ops/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: lat_fs -s 10K /ext2
+  legend: Number of created/deleted files on {system}
+  title: '[Ext2] The throughput of creating/deleting small files (10KB)'
+  unit: number
+result_extraction:
+  result_index: 2
+  search_pattern: ^0k\t[0-9]+
diff --git a/test/benchmark/lmbench/fifo_lat/bench_result.json b/test/benchmark/lmbench/fifo_lat/bench_result.json
deleted file mode 100644
index 8f9dd75d..00000000
--- a/test/benchmark/lmbench/fifo_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Fifo latency: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[FIFO] The cost of write+read (1B)",
-        "description": "lat_fifo",
-        "unit": "\u00b5s",
-        "legend": "Average fifo latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/fifo_lat/bench_result.yaml b/test/benchmark/lmbench/fifo_lat/bench_result.yaml
new file mode 100644
index 00000000..9c4c8f45
--- /dev/null
+++ b/test/benchmark/lmbench/fifo_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_fifo
+  legend: Average fifo latency on {system}
+  title: '[FIFO] The cost of write+read (1B)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Fifo latency: +[0-9.]+'
diff --git a/test/benchmark/lmbench/mem_copy_bw/bench_result.json b/test/benchmark/lmbench/mem_copy_bw/bench_result.json
deleted file mode 100644
index 14d3b8b5..00000000
--- a/test/benchmark/lmbench/mem_copy_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "536.87 +[0-9.]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Memory] The bandwidth of copying integers",
-        "description": "bw_mem fcp",
-        "unit": "MB/s",
-        "legend": "Average memory copy bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/mem_copy_bw/bench_result.yaml b/test/benchmark/lmbench/mem_copy_bw/bench_result.yaml
new file mode 100644
index 00000000..2d69c47e
--- /dev/null
+++ b/test/benchmark/lmbench/mem_copy_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_mem fcp
+  legend: Average memory copy bandwidth on {system}
+  title: '[Memory] The bandwidth of copying integers'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: 536.87 +[0-9.]+
diff --git a/test/benchmark/lmbench/mem_mmap_bw/bench_result.json b/test/benchmark/lmbench/mem_mmap_bw/bench_result.json
deleted file mode 100644
index 66c5f510..00000000
--- a/test/benchmark/lmbench/mem_mmap_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "268.44 +[0-9.]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Memory] The bandwidth of mmap",
-        "description": "bw_mmap",
-        "unit": "MB/s",
-        "legend": "Average mmap bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/mem_mmap_bw/bench_result.yaml b/test/benchmark/lmbench/mem_mmap_bw/bench_result.yaml
new file mode 100644
index 00000000..95b0096d
--- /dev/null
+++ b/test/benchmark/lmbench/mem_mmap_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_mmap
+  legend: Average mmap bandwidth on {system}
+  title: '[Memory] The bandwidth of mmap'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: 268.44 +[0-9.]+
diff --git a/test/benchmark/lmbench/mem_mmap_lat/bench_result.json b/test/benchmark/lmbench/mem_mmap_lat/bench_result.json
deleted file mode 100644
index 4181e4d9..00000000
--- a/test/benchmark/lmbench/mem_mmap_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "4.194304 +[0-9.]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Memory] The cost of mmap+unmap",
-        "description": "lat_mmap",
-        "unit": "\u00b5s",
-        "legend": "Average mmap latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/mem_mmap_lat/bench_result.yaml b/test/benchmark/lmbench/mem_mmap_lat/bench_result.yaml
new file mode 100644
index 00000000..17459ba4
--- /dev/null
+++ b/test/benchmark/lmbench/mem_mmap_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_mmap
+  legend: Average mmap latency on {system}
+  title: '[Memory] The cost of mmap+unmap'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 2
+  search_pattern: 4.194304 +[0-9.]+
diff --git a/test/benchmark/lmbench/mem_pagefault_lat/bench_result.json b/test/benchmark/lmbench/mem_pagefault_lat/bench_result.json
deleted file mode 100644
index 456d45bf..00000000
--- a/test/benchmark/lmbench/mem_pagefault_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Pagefaults on [^:]+: +[0-9.]+",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "[Memory] The cost of page fault handling",
-        "description": "lat_pagefault",
-        "unit": "\u00b5s",
-        "legend": "Average page fault latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/mem_pagefault_lat/bench_result.yaml b/test/benchmark/lmbench/mem_pagefault_lat/bench_result.yaml
new file mode 100644
index 00000000..52cba806
--- /dev/null
+++ b/test/benchmark/lmbench/mem_pagefault_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_pagefault
+  legend: Average page fault latency on {system}
+  title: '[Memory] The cost of page fault handling'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 4
+  search_pattern: 'Pagefaults on [^:]+: +[0-9.]+'
diff --git a/test/benchmark/lmbench/mem_read_bw/bench_result.json b/test/benchmark/lmbench/mem_read_bw/bench_result.json
deleted file mode 100644
index c5968678..00000000
--- a/test/benchmark/lmbench/mem_read_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "536.87 +[0-9.]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Memory] The bandwidth of reading integers",
-        "description": "bw_mem frd",
-        "unit": "MB/s",
-        "legend": "Average memory read bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/mem_read_bw/bench_result.yaml b/test/benchmark/lmbench/mem_read_bw/bench_result.yaml
new file mode 100644
index 00000000..1c694bbf
--- /dev/null
+++ b/test/benchmark/lmbench/mem_read_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_mem frd
+  legend: Average memory read bandwidth on {system}
+  title: '[Memory] The bandwidth of reading integers'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: 536.87 +[0-9.]+
diff --git a/test/benchmark/lmbench/mem_write_bw/bench_result.json b/test/benchmark/lmbench/mem_write_bw/bench_result.json
deleted file mode 100644
index 5c31b523..00000000
--- a/test/benchmark/lmbench/mem_write_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "536.87 +[0-9.]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Memory] The bandwidth of writing integers",
-        "description": "bw_mem fwr",
-        "unit": "MB/s",
-        "legend": "Average memory write bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/mem_write_bw/bench_result.yaml b/test/benchmark/lmbench/mem_write_bw/bench_result.yaml
new file mode 100644
index 00000000..b41fab7d
--- /dev/null
+++ b/test/benchmark/lmbench/mem_write_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_mem fwr
+  legend: Average memory write bandwidth on {system}
+  title: '[Memory] The bandwidth of writing integers'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: 536.87 +[0-9.]+
diff --git a/test/benchmark/lmbench/pipe_bw/bench_result.json b/test/benchmark/lmbench/pipe_bw/bench_result.json
deleted file mode 100644
index 5c219611..00000000
--- a/test/benchmark/lmbench/pipe_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Pipe bandwidth: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Pipes] The bandwidth",
-        "description": "bw_pipe",
-        "unit": "MB/s",
-        "legend": "Average pipe bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/pipe_bw/bench_result.yaml b/test/benchmark/lmbench/pipe_bw/bench_result.yaml
new file mode 100644
index 00000000..f6de3737
--- /dev/null
+++ b/test/benchmark/lmbench/pipe_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_pipe
+  legend: Average pipe bandwidth on {system}
+  title: '[Pipes] The bandwidth'
+  unit: MB/s
+result_extraction:
+  result_index: 3
+  search_pattern: 'Pipe bandwidth: +[0-9.]+'
diff --git a/test/benchmark/lmbench/pipe_lat/bench_result.json b/test/benchmark/lmbench/pipe_lat/bench_result.json
deleted file mode 100644
index 172e9a03..00000000
--- a/test/benchmark/lmbench/pipe_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Pipe latency: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Pipes] The cost of write+read (1B)",
-        "description": "lat_pipe",
-        "unit": "\u00b5s",
-        "legend": "Average pipe latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/pipe_lat/bench_result.yaml b/test/benchmark/lmbench/pipe_lat/bench_result.yaml
new file mode 100644
index 00000000..495e957a
--- /dev/null
+++ b/test/benchmark/lmbench/pipe_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_pipe
+  legend: Average pipe latency on {system}
+  title: '[Pipes] The cost of write+read (1B)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Pipe latency: +[0-9.]+'
diff --git a/test/benchmark/lmbench/process_ctx_lat/bench_result.json b/test/benchmark/lmbench/process_ctx_lat/bench_result.json
deleted file mode 100644
index b83cef6a..00000000
--- a/test/benchmark/lmbench/process_ctx_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "18 +[0-9.]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Process] The cost of context switching",
-        "description": "lat_ctx 2",
-        "unit": "\u00b5s",
-        "legend": "Average context switch latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/process_ctx_lat/bench_result.yaml b/test/benchmark/lmbench/process_ctx_lat/bench_result.yaml
new file mode 100644
index 00000000..0d1e1e04
--- /dev/null
+++ b/test/benchmark/lmbench/process_ctx_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_ctx 2
+  legend: Average context switch latency on {system}
+  title: '[Process] The cost of context switching'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 2
+  search_pattern: 18 +[0-9.]+
diff --git a/test/benchmark/lmbench/process_exec_lat/bench_result.json b/test/benchmark/lmbench/process_exec_lat/bench_result.json
deleted file mode 100644
index 9015ccfa..00000000
--- a/test/benchmark/lmbench/process_exec_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Process fork\\+execve: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Process] The cost of fork+exec+exit",
-        "description": "lat_proc exec",
-        "unit": "\u00b5s",
-        "legend": "Average exec latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/process_exec_lat/bench_result.yaml b/test/benchmark/lmbench/process_exec_lat/bench_result.yaml
new file mode 100644
index 00000000..e71b080d
--- /dev/null
+++ b/test/benchmark/lmbench/process_exec_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_proc exec
+  legend: Average exec latency on {system}
+  title: '[Process] The cost of fork+exec+exit'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Process fork\+execve: +[0-9.]+'
diff --git a/test/benchmark/lmbench/process_fork_lat/bench_result.json b/test/benchmark/lmbench/process_fork_lat/bench_result.json
deleted file mode 100644
index fa69c0d9..00000000
--- a/test/benchmark/lmbench/process_fork_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Process fork\\+exit: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Process] The cost of fork+exit",
-        "description": "lat_proc fork",
-        "unit": "\u00b5s",
-        "legend": "Average Fork latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/process_fork_lat/bench_result.yaml b/test/benchmark/lmbench/process_fork_lat/bench_result.yaml
new file mode 100644
index 00000000..501a0529
--- /dev/null
+++ b/test/benchmark/lmbench/process_fork_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_proc fork
+  legend: Average Fork latency on {system}
+  title: '[Process] The cost of fork+exit'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Process fork\+exit: +[0-9.]+'
diff --git a/test/benchmark/lmbench/process_getppid_lat/bench_result.json b/test/benchmark/lmbench/process_getppid_lat/bench_result.json
deleted file mode 100644
index 254fabbb..00000000
--- a/test/benchmark/lmbench/process_getppid_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Simple syscall: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Process] The cost of getppid",
-        "description": "lat_syscall null",
-        "unit": "\u00b5s",
-        "legend": "Average syscall latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/process_getppid_lat/bench_result.yaml b/test/benchmark/lmbench/process_getppid_lat/bench_result.yaml
new file mode 100644
index 00000000..2998840e
--- /dev/null
+++ b/test/benchmark/lmbench/process_getppid_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_syscall null
+  legend: Average syscall latency on {system}
+  title: '[Process] The cost of getppid'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Simple syscall: +[0-9.]+'
diff --git a/test/benchmark/lmbench/process_shell_lat/bench_result.json b/test/benchmark/lmbench/process_shell_lat/bench_result.json
deleted file mode 100644
index e2c712fd..00000000
--- a/test/benchmark/lmbench/process_shell_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Process fork\\+\\/bin\\/sh",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "[Process] The cost of fork+exec+shell+exit",
-        "description": "lat_proc shell",
-        "unit": "\u00b5s",
-        "legend": "Average shell latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/process_shell_lat/bench_result.yaml b/test/benchmark/lmbench/process_shell_lat/bench_result.yaml
new file mode 100644
index 00000000..d32d7a00
--- /dev/null
+++ b/test/benchmark/lmbench/process_shell_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_proc shell
+  legend: Average shell latency on {system}
+  title: '[Process] The cost of fork+exec+shell+exit'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 4
+  search_pattern: Process fork\+\/bin\/sh
diff --git a/test/benchmark/lmbench/ramfs_copy_files_bw/bench_result.json b/test/benchmark/lmbench/ramfs_copy_files_bw/bench_result.json
deleted file mode 100644
index d6301ecb..00000000
--- a/test/benchmark/lmbench/ramfs_copy_files_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "lmdd result:",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ramfs] The bandwidth of copying data between files",
-        "description": "lmdd",
-        "unit": "MB/s",
-        "legend": "Average file copy bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/ramfs_copy_files_bw/bench_result.yaml b/test/benchmark/lmbench/ramfs_copy_files_bw/bench_result.yaml
new file mode 100644
index 00000000..3b2ce273
--- /dev/null
+++ b/test/benchmark/lmbench/ramfs_copy_files_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: lmdd
+  legend: Average file copy bandwidth on {system}
+  title: '[Ramfs] The bandwidth of copying data between files'
+  unit: MB/s
+result_extraction:
+  result_index: 8
+  search_pattern: 'lmdd result:'
diff --git a/test/benchmark/lmbench/ramfs_create_delete_files_0k_ops/bench_result.json b/test/benchmark/lmbench/ramfs_create_delete_files_0k_ops/bench_result.json
deleted file mode 100644
index 601d93bd..00000000
--- a/test/benchmark/lmbench/ramfs_create_delete_files_0k_ops/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "^0k\t[0-9]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Ramfs] The throughput of creating/deleting small files (0KB)",
-        "description": "lat_fs -s 0k",
-        "unit": "number",
-        "legend": "Number of created/deleted files on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/ramfs_create_delete_files_0k_ops/bench_result.yaml b/test/benchmark/lmbench/ramfs_create_delete_files_0k_ops/bench_result.yaml
new file mode 100644
index 00000000..8ece993f
--- /dev/null
+++ b/test/benchmark/lmbench/ramfs_create_delete_files_0k_ops/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: lat_fs -s 0k
+  legend: Number of created/deleted files on {system}
+  title: '[Ramfs] The throughput of creating/deleting small files (0KB)'
+  unit: number
+result_extraction:
+  result_index: 2
+  search_pattern: ^0k
diff --git a/test/benchmark/lmbench/ramfs_create_delete_files_10k_ops/bench_result.json b/test/benchmark/lmbench/ramfs_create_delete_files_10k_ops/bench_result.json
deleted file mode 100644
index c4e3caba..00000000
--- a/test/benchmark/lmbench/ramfs_create_delete_files_10k_ops/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "^10k\t[0-9]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Ramfs] The throughput of creating/deleting small files (10KB)",
-        "description": "lat_fs -s 10K",
-        "unit": "number",
-        "legend": "Number of created/deleted files on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/ramfs_create_delete_files_10k_ops/bench_result.yaml b/test/benchmark/lmbench/ramfs_create_delete_files_10k_ops/bench_result.yaml
new file mode 100644
index 00000000..3a685a01
--- /dev/null
+++ b/test/benchmark/lmbench/ramfs_create_delete_files_10k_ops/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: lat_fs -s 10K
+  legend: Number of created/deleted files on {system}
+  title: '[Ramfs] The throughput of creating/deleting small files (10KB)'
+  unit: number
+result_extraction:
+  result_index: 2
+  search_pattern: 10k
diff --git a/test/benchmark/lmbench/semaphore_lat/bench_result.json b/test/benchmark/lmbench/semaphore_lat/bench_result.json
deleted file mode 100644
index 47800203..00000000
--- a/test/benchmark/lmbench/semaphore_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Semaphore latency: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Semaphores] The cost of semop",
-        "description": "lat_sem",
-        "unit": "\u00b5s",
-        "legend": "Average semaphore latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/semaphore_lat/bench_result.yaml b/test/benchmark/lmbench/semaphore_lat/bench_result.yaml
new file mode 100644
index 00000000..c85d4cff
--- /dev/null
+++ b/test/benchmark/lmbench/semaphore_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_sem
+  legend: Average semaphore latency on {system}
+  title: '[Semaphores] The cost of semop'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Semaphore latency: +[0-9.]+'
diff --git a/test/benchmark/lmbench/signal_catch_lat/bench_result.json b/test/benchmark/lmbench/signal_catch_lat/bench_result.json
deleted file mode 100644
index ff9ea207..00000000
--- a/test/benchmark/lmbench/signal_catch_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Signal handler overhead: +[0-9.]+",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "[Signals] The cost of catching a signal",
-        "description": "lat_sig catch",
-        "unit": "\u00b5s",
-        "legend": "Average Signal handler overhead on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/signal_catch_lat/bench_result.yaml b/test/benchmark/lmbench/signal_catch_lat/bench_result.yaml
new file mode 100644
index 00000000..9466b8ad
--- /dev/null
+++ b/test/benchmark/lmbench/signal_catch_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_sig catch
+  legend: Average Signal handler overhead on {system}
+  title: '[Signals] The cost of catching a signal'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 4
+  search_pattern: 'Signal handler overhead: +[0-9.]+'
diff --git a/test/benchmark/lmbench/signal_install_lat/bench_result.json b/test/benchmark/lmbench/signal_install_lat/bench_result.json
deleted file mode 100644
index a926bd98..00000000
--- a/test/benchmark/lmbench/signal_install_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Signal handler installation: +[0-9.]+",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "[Signals] The cost of installing a signal handler",
-        "description": "lat_sig install",
-        "unit": "\u00b5s",
-        "legend": "Average Signal handler install latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/signal_install_lat/bench_result.yaml b/test/benchmark/lmbench/signal_install_lat/bench_result.yaml
new file mode 100644
index 00000000..171ed8b2
--- /dev/null
+++ b/test/benchmark/lmbench/signal_install_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_sig install
+  legend: Average Signal handler install latency on {system}
+  title: '[Signals] The cost of installing a signal handler'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 4
+  search_pattern: 'Signal handler installation: +[0-9.]+'
diff --git a/test/benchmark/lmbench/signal_prot_lat/bench_result.json b/test/benchmark/lmbench/signal_prot_lat/bench_result.json
deleted file mode 100644
index 44868122..00000000
--- a/test/benchmark/lmbench/signal_prot_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Protection fault: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Signals] The cost of catching a segfault",
-        "description": "lat_sig prot",
-        "unit": "\u00b5s",
-        "legend": "Average protection fault latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/signal_prot_lat/bench_result.yaml b/test/benchmark/lmbench/signal_prot_lat/bench_result.yaml
new file mode 100644
index 00000000..55707a31
--- /dev/null
+++ b/test/benchmark/lmbench/signal_prot_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_sig prot
+  legend: Average protection fault latency on {system}
+  title: '[Signals] The cost of catching a segfault'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Protection fault: +[0-9.]+'
diff --git a/test/benchmark/lmbench/summary.json b/test/benchmark/lmbench/summary.json
deleted file mode 100644
index 53a1bd5c..00000000
--- a/test/benchmark/lmbench/summary.json
+++ /dev/null
@@ -1,44 +0,0 @@
-{
-    "benchmarks": [
-        "mem_pagefault_lat",
-        "mem_mmap_bw",
-        "mem_mmap_lat",
-        "process_getppid_lat",
-        "process_ctx_lat",
-        "process_fork_lat",
-        "process_exec_lat",
-        "process_shell_lat",
-        "signal_install_lat",
-        "signal_catch_lat",
-        "signal_prot_lat",
-        "pipe_lat",
-        "pipe_bw",
-        "fifo_lat",
-        "unix_lat",
-        "unix_bw",
-        "unix_connect_lat",
-        "vfs_open_lat",
-        "vfs_read_lat",
-        "vfs_write_lat",
-        "vfs_stat_lat",
-        "vfs_fstat_lat",
-        "vfs_read_pagecache_bw",
-        "vfs_select_lat",
-        "ramfs_copy_files_bw",
-        "ramfs_create_delete_files_0k_ops",
-        "ramfs_create_delete_files_10k_ops",
-        "ext2_copy_files_bw",
-        "tcp_loopback_bw_128",
-        "tcp_loopback_bw_4k",
-        "tcp_loopback_bw_64k",
-        "tcp_loopback_lat",
-        "tcp_loopback_select_lat",
-        "tcp_loopback_http_bw",
-        "udp_loopback_lat",
-        "tcp_virtio_bw_128",
-        "tcp_virtio_bw_64k",
-        "tcp_virtio_connect_lat",
-        "tcp_virtio_lat",
-        "udp_virtio_lat"
-    ]
-}
diff --git a/test/benchmark/lmbench/summary.yaml b/test/benchmark/lmbench/summary.yaml
new file mode 100644
index 00000000..b6fd6b04
--- /dev/null
+++ b/test/benchmark/lmbench/summary.yaml
@@ -0,0 +1,45 @@
+benchmarks:
+  - mem_pagefault_lat
+  - mem_mmap_bw
+  - mem_mmap_lat
+  - process_getppid_lat
+  - process_ctx_lat
+  - process_fork_lat
+  - process_exec_lat
+  - process_shell_lat
+  - signal_install_lat
+  - signal_catch_lat
+  - signal_prot_lat
+  - pipe_lat
+  - pipe_bw
+  - fifo_lat
+  - unix_lat
+  - unix_bw
+  - unix_connect_lat
+  - vfs_open_lat
+  - vfs_read_lat
+  - vfs_write_lat
+  - vfs_stat_lat
+  - vfs_fstat_lat
+  - vfs_read_pagecache_bw
+  - vfs_select_lat
+  - ramfs_copy_files_bw
+  - ramfs_create_delete_files_0k_ops
+  - ramfs_create_delete_files_10k_ops
+  - ext2_copy_files_bw
+  - tcp_loopback_bw_128
+  - tcp_loopback_bw_4k
+  - tcp_loopback_bw_64k
+  - tcp_loopback_lat
+  - tcp_loopback_select_lat
+  - tcp_loopback_http_bw
+  - udp_loopback_lat
+  - tcp_virtio_bw_128
+  - tcp_virtio_bw_64k
+  - tcp_virtio_connect_lat
+  - tcp_virtio_lat
+  - udp_virtio_lat
+  - # mem_copy_bw
+  - # mem_read_bw
+  - # mem_write_bw
+  - # vfs_fcntl_lat
\ No newline at end of file
diff --git a/test/benchmark/lmbench/tcp_loopback_bw_128/bench_result.json b/test/benchmark/lmbench/tcp_loopback_bw_128/bench_result.json
deleted file mode 100644
index 42250d09..00000000
--- a/test/benchmark/lmbench/tcp_loopback_bw_128/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "0.000128 +[0-9.]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[TCP sockets] The bandwidth (loopback, 128B message)",
-        "description": "bw_tcp -l",
-        "unit": "MB/s",
-        "legend": "Average TCP bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/tcp_loopback_bw_128/bench_result.yaml b/test/benchmark/lmbench/tcp_loopback_bw_128/bench_result.yaml
new file mode 100644
index 00000000..47b6fbc8
--- /dev/null
+++ b/test/benchmark/lmbench/tcp_loopback_bw_128/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_tcp -l
+  legend: Average TCP bandwidth on {system}
+  title: '[TCP sockets] The bandwidth (loopback, 128B message)'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: 0.000128 +[0-9.]+
diff --git a/test/benchmark/lmbench/tcp_loopback_bw_4k/bench_result.json b/test/benchmark/lmbench/tcp_loopback_bw_4k/bench_result.json
deleted file mode 100644
index 584cb673..00000000
--- a/test/benchmark/lmbench/tcp_loopback_bw_4k/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "0.004096 +[0-9.]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[TCP sockets] The bandwidth (loopback, 4KB message)",
-        "description": "bw_tcp -l",
-        "unit": "MB/s",
-        "legend": "Average TCP bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/tcp_loopback_bw_4k/bench_result.yaml b/test/benchmark/lmbench/tcp_loopback_bw_4k/bench_result.yaml
new file mode 100644
index 00000000..a6aec939
--- /dev/null
+++ b/test/benchmark/lmbench/tcp_loopback_bw_4k/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_tcp -l
+  legend: Average TCP bandwidth on {system}
+  title: '[TCP sockets] The bandwidth (loopback, 4KB message)'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: 0.004096 +[0-9.]+
diff --git a/test/benchmark/lmbench/tcp_loopback_bw_64k/bench_result.json b/test/benchmark/lmbench/tcp_loopback_bw_64k/bench_result.json
deleted file mode 100644
index 2dedc33a..00000000
--- a/test/benchmark/lmbench/tcp_loopback_bw_64k/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "0.065536 +[0-9.]+ +MB",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[TCP sockets] The bandwidth (loopback, 64KB message)",
-        "description": "bw_tcp -l",
-        "unit": "MB/s",
-        "legend": "Average TCP bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/tcp_loopback_bw_64k/bench_result.yaml b/test/benchmark/lmbench/tcp_loopback_bw_64k/bench_result.yaml
new file mode 100644
index 00000000..56216f82
--- /dev/null
+++ b/test/benchmark/lmbench/tcp_loopback_bw_64k/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_tcp -l
+  legend: Average TCP bandwidth on {system}
+  title: '[TCP sockets] The bandwidth (loopback, 64KB message)'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: 0.065536 +[0-9.]+ +MB
diff --git a/test/benchmark/lmbench/tcp_loopback_connect_lat/bench_result.json b/test/benchmark/lmbench/tcp_loopback_connect_lat/bench_result.json
deleted file mode 100644
index df84606a..00000000
--- a/test/benchmark/lmbench/tcp_loopback_connect_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "TCP\\/IP connection cost to [0-9.]+: +[0-9.]+",
-        "result_index": 6
-    },
-    "chart": {
-        "title": "[TCP sockets] The latency of connect (loopback)",
-        "description": "lat_connect",
-        "unit": "\u00b5s",
-        "legend": "Average TCP connection latency on {system}"
-    }
-}
diff --git a/test/benchmark/lmbench/tcp_loopback_connect_lat/bench_result.yaml b/test/benchmark/lmbench/tcp_loopback_connect_lat/bench_result.yaml
new file mode 100644
index 00000000..b5834630
--- /dev/null
+++ b/test/benchmark/lmbench/tcp_loopback_connect_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_connect
+  legend: Average TCP connection latency on {system}
+  title: '[TCP sockets] The latency of connect (loopback)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 6
+  search_pattern: 'TCP\/IP connection cost to [0-9.]+: +[0-9.]+'
diff --git a/test/benchmark/lmbench/tcp_loopback_http_bw/bench_result.json b/test/benchmark/lmbench/tcp_loopback_http_bw/bench_result.json
deleted file mode 100644
index 95f4ee34..00000000
--- a/test/benchmark/lmbench/tcp_loopback_http_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Avg xfer: +([0-9.]+|inf)",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[HTTP] The bandwidth (loopback)",
-        "description": "bw_http",
-        "unit": "MB/s",
-        "legend": "Average simple HTTP transaction bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/tcp_loopback_http_bw/bench_result.yaml b/test/benchmark/lmbench/tcp_loopback_http_bw/bench_result.yaml
new file mode 100644
index 00000000..ab250a17
--- /dev/null
+++ b/test/benchmark/lmbench/tcp_loopback_http_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_http
+  legend: Average simple HTTP transaction bandwidth on {system}
+  title: '[HTTP] The bandwidth (loopback)'
+  unit: MB/s
+result_extraction:
+  result_index: 8
+  search_pattern: 'Avg xfer: +([0-9.]+|inf)'
diff --git a/test/benchmark/lmbench/tcp_loopback_lat/bench_result.json b/test/benchmark/lmbench/tcp_loopback_lat/bench_result.json
deleted file mode 100644
index 07f9d80c..00000000
--- a/test/benchmark/lmbench/tcp_loopback_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "TCP latency using [0-9.]+: +[0-9.]+",
-        "result_index": 5
-    },
-    "chart": {
-        "title": "[TCP sockets] The latency of write+read (loopback, 1B message)",
-        "description": "lat_tcp",
-        "unit": "\u00b5s",
-        "legend": "Average TCP latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/tcp_loopback_lat/bench_result.yaml b/test/benchmark/lmbench/tcp_loopback_lat/bench_result.yaml
new file mode 100644
index 00000000..acda2f70
--- /dev/null
+++ b/test/benchmark/lmbench/tcp_loopback_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_tcp
+  legend: Average TCP latency on {system}
+  title: '[TCP sockets] The latency of write+read (loopback, 1B message)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 5
+  search_pattern: 'TCP latency using [0-9.]+: +[0-9.]+'
diff --git a/test/benchmark/lmbench/tcp_loopback_select_lat/bench_result.json b/test/benchmark/lmbench/tcp_loopback_select_lat/bench_result.json
deleted file mode 100644
index 12e0b6b9..00000000
--- a/test/benchmark/lmbench/tcp_loopback_select_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Select on [0-9.]+ tcp fd's: +[0-9.]+",
-        "result_index": 6
-    },
-    "chart": {
-        "title": "[Network] The cost of select (TCP fds)",
-        "description": "lat_select",
-        "unit": "\u00b5s",
-        "legend": "Average select TCP latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/tcp_loopback_select_lat/bench_result.yaml b/test/benchmark/lmbench/tcp_loopback_select_lat/bench_result.yaml
new file mode 100644
index 00000000..91878221
--- /dev/null
+++ b/test/benchmark/lmbench/tcp_loopback_select_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_select
+  legend: Average select TCP latency on {system}
+  title: '[Network] The cost of select (TCP fds)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 6
+  search_pattern: 'Select on [0-9.]+ tcp fd''s: +[0-9.]+'
diff --git a/test/benchmark/lmbench/tcp_virtio_bw_128/bench_result.json b/test/benchmark/lmbench/tcp_virtio_bw_128/bench_result.json
deleted file mode 100644
index bdcca99b..00000000
--- a/test/benchmark/lmbench/tcp_virtio_bw_128/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "0.000128 +[0-9.]+ +MB",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[TCP sockets] The bandwidth (virtio-net, 128B message)",
-        "description": "bw_tcp -l",
-        "unit": "MB/sec",
-        "legend": "Average TCP bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/tcp_virtio_bw_128/bench_result.yaml b/test/benchmark/lmbench/tcp_virtio_bw_128/bench_result.yaml
new file mode 100644
index 00000000..9eefc955
--- /dev/null
+++ b/test/benchmark/lmbench/tcp_virtio_bw_128/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_tcp -l
+  legend: Average TCP bandwidth on {system}
+  title: '[TCP sockets] The bandwidth (virtio-net, 128B message)'
+  unit: MB/sec
+result_extraction:
+  result_index: 2
+  search_pattern: 0.000128 +[0-9.]+ +MB
diff --git a/test/benchmark/lmbench/tcp_virtio_bw_64k/bench_result.json b/test/benchmark/lmbench/tcp_virtio_bw_64k/bench_result.json
deleted file mode 100644
index 42a2f641..00000000
--- a/test/benchmark/lmbench/tcp_virtio_bw_64k/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "0.065536 +[0-9.]+ +MB",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[TCP sockets] The bandwidth (virtio-net, 64KB message)",
-        "description": "bw_tcp -l",
-        "unit": "MB/sec",
-        "legend": "Average TCP bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/tcp_virtio_bw_64k/bench_result.yaml b/test/benchmark/lmbench/tcp_virtio_bw_64k/bench_result.yaml
new file mode 100644
index 00000000..774cca8b
--- /dev/null
+++ b/test/benchmark/lmbench/tcp_virtio_bw_64k/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_tcp -l
+  legend: Average TCP bandwidth on {system}
+  title: '[TCP sockets] The bandwidth (virtio-net, 64KB message)'
+  unit: MB/sec
+result_extraction:
+  result_index: 2
+  search_pattern: 0.065536 +[0-9.]+ +MB
diff --git a/test/benchmark/lmbench/tcp_virtio_connect_lat/bench_result.json b/test/benchmark/lmbench/tcp_virtio_connect_lat/bench_result.json
deleted file mode 100644
index c020f529..00000000
--- a/test/benchmark/lmbench/tcp_virtio_connect_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "TCP\\/IP connection cost to [0-9.]+: +[0-9.]+",
-        "result_index": 6
-    },
-    "chart": {
-        "title": "[TCP sockets] The latency of connect (virtio-net)",
-        "description": "lat_connect",
-        "unit": "\u00b5s",
-        "legend": "Average TCP connection latency on {system}"
-    }
-}
diff --git a/test/benchmark/lmbench/tcp_virtio_connect_lat/bench_result.yaml b/test/benchmark/lmbench/tcp_virtio_connect_lat/bench_result.yaml
new file mode 100644
index 00000000..0cb8ed06
--- /dev/null
+++ b/test/benchmark/lmbench/tcp_virtio_connect_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_connect
+  legend: Average TCP connection latency on {system}
+  title: '[TCP sockets] The latency of connect (virtio-net)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 6
+  search_pattern: 'TCP\/IP connection cost to [0-9.]+: +[0-9.]+'
diff --git a/test/benchmark/lmbench/tcp_virtio_lat/bench_result.json b/test/benchmark/lmbench/tcp_virtio_lat/bench_result.json
deleted file mode 100644
index 876dfe53..00000000
--- a/test/benchmark/lmbench/tcp_virtio_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "TCP latency using [0-9.]+: +[0-9.]+",
-        "result_index": 5
-    },
-    "chart": {
-        "title": "[TCP sockets] The latency of write+read (virtio-net, 1B message)",
-        "description": "lat_tcp_virtio",
-        "unit": "\u00b5s",
-        "legend": "Average TCP latency over virtio-net on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/tcp_virtio_lat/bench_result.yaml b/test/benchmark/lmbench/tcp_virtio_lat/bench_result.yaml
new file mode 100644
index 00000000..3568eee4
--- /dev/null
+++ b/test/benchmark/lmbench/tcp_virtio_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_tcp_virtio
+  legend: Average TCP latency over virtio-net on {system}
+  title: '[TCP sockets] The latency of write+read (virtio-net, 1B message)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 5
+  search_pattern: 'TCP latency using [0-9.]+: +[0-9.]+'
diff --git a/test/benchmark/lmbench/udp_loopback_lat/bench_result.json b/test/benchmark/lmbench/udp_loopback_lat/bench_result.json
deleted file mode 100644
index 2a3bb8d4..00000000
--- a/test/benchmark/lmbench/udp_loopback_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "UDP latency using [0-9.]+: +[0-9.]+",
-        "result_index": 5
-    },
-    "chart": {
-        "title": "[UDP sockets] The latency of write+read (loopback, 4B message)",
-        "description": "lat_udp",
-        "unit": "\u00b5s",
-        "legend": "Average UDP latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/udp_loopback_lat/bench_result.yaml b/test/benchmark/lmbench/udp_loopback_lat/bench_result.yaml
new file mode 100644
index 00000000..5c03bbfc
--- /dev/null
+++ b/test/benchmark/lmbench/udp_loopback_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_udp
+  legend: Average UDP latency on {system}
+  title: '[UDP sockets] The latency of write+read (loopback, 4B message)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 5
+  search_pattern: 'UDP latency using [0-9.]+: +[0-9.]+'
diff --git a/test/benchmark/lmbench/udp_virtio_lat/bench_result.json b/test/benchmark/lmbench/udp_virtio_lat/bench_result.json
deleted file mode 100644
index 94ca7aa6..00000000
--- a/test/benchmark/lmbench/udp_virtio_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "UDP latency using [0-9.]+: +[0-9.]+",
-        "result_index": 5
-    },
-    "chart": {
-        "title": "[UDP sockets] The latency of write+read (virtio-net, 4B message)",
-        "description": "lat_udp",
-        "unit": "\u00b5s",
-        "legend": "Average UDP latency over virtio-net on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/udp_virtio_lat/bench_result.yaml b/test/benchmark/lmbench/udp_virtio_lat/bench_result.yaml
new file mode 100644
index 00000000..c50b66a6
--- /dev/null
+++ b/test/benchmark/lmbench/udp_virtio_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_udp
+  legend: Average UDP latency over virtio-net on {system}
+  title: '[UDP sockets] The latency of write+read (virtio-net, 4B message)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 5
+  search_pattern: 'UDP latency using [0-9.]+: +[0-9.]+'
diff --git a/test/benchmark/lmbench/unix_bw/bench_result.json b/test/benchmark/lmbench/unix_bw/bench_result.json
deleted file mode 100644
index 0c2dfea5..00000000
--- a/test/benchmark/lmbench/unix_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "sock stream bandwidth: +[0-9.]+",
-        "result_index": 5
-    },
-    "chart": {
-        "title": "[Unix sockets] The bandwidth (64KB message)",
-        "description": "bw_unix",
-        "unit": "MB/s",
-        "legend": "Average unix bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/unix_bw/bench_result.yaml b/test/benchmark/lmbench/unix_bw/bench_result.yaml
new file mode 100644
index 00000000..0239d0f6
--- /dev/null
+++ b/test/benchmark/lmbench/unix_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_unix
+  legend: Average unix bandwidth on {system}
+  title: '[Unix sockets] The bandwidth (64KB message)'
+  unit: MB/s
+result_extraction:
+  result_index: 5
+  search_pattern: 'sock stream bandwidth: +[0-9.]+'
diff --git a/test/benchmark/lmbench/unix_connect_lat/bench_result.json b/test/benchmark/lmbench/unix_connect_lat/bench_result.json
deleted file mode 100644
index 24aa6598..00000000
--- a/test/benchmark/lmbench/unix_connect_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "UNIX connection cost: +[0-9.]+",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "[Unix sockets] The latency of connect",
-        "description": "lat_connect",
-        "unit": "\u00b5s",
-        "legend": "Average unix connect latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/unix_connect_lat/bench_result.yaml b/test/benchmark/lmbench/unix_connect_lat/bench_result.yaml
new file mode 100644
index 00000000..f5ca6b88
--- /dev/null
+++ b/test/benchmark/lmbench/unix_connect_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_connect
+  legend: Average unix connect latency on {system}
+  title: '[Unix sockets] The latency of connect'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 4
+  search_pattern: 'UNIX connection cost: +[0-9.]+'
diff --git a/test/benchmark/lmbench/unix_lat/bench_result.json b/test/benchmark/lmbench/unix_lat/bench_result.json
deleted file mode 100644
index 21291db2..00000000
--- a/test/benchmark/lmbench/unix_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "sock stream latency: +[0-9.]+",
-        "result_index": 5
-    },
-    "chart": {
-        "title": "[Unix sockets] The latency of write+read (1B message)",
-        "description": "lat_unix",
-        "unit": "\u00b5s",
-        "legend": "Average unix latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/unix_lat/bench_result.yaml b/test/benchmark/lmbench/unix_lat/bench_result.yaml
new file mode 100644
index 00000000..0536a2c7
--- /dev/null
+++ b/test/benchmark/lmbench/unix_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_unix
+  legend: Average unix latency on {system}
+  title: '[Unix sockets] The latency of write+read (1B message)'
+  unit: "\xB5s"
+result_extraction:
+  result_index: 5
+  search_pattern: 'sock stream latency: +[0-9.]+'
diff --git a/test/benchmark/lmbench/vfs_fcntl_lat/bench_result.json b/test/benchmark/lmbench/vfs_fcntl_lat/bench_result.json
deleted file mode 100644
index 40c3eefa..00000000
--- a/test/benchmark/lmbench/vfs_fcntl_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Fcntl lock latency: +[0-9.]+",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "[VFS] The cost of record locking/unlocking via fcntl",
-        "description": "lat_fcntl",
-        "unit": "\u00b5s",
-        "legend": "Average file locking latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/vfs_fcntl_lat/bench_result.yaml b/test/benchmark/lmbench/vfs_fcntl_lat/bench_result.yaml
new file mode 100644
index 00000000..dccbbfc5
--- /dev/null
+++ b/test/benchmark/lmbench/vfs_fcntl_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_fcntl
+  legend: Average file locking latency on {system}
+  title: '[VFS] The cost of record locking/unlocking via fcntl'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 4
+  search_pattern: 'Fcntl lock latency: +[0-9.]+'
diff --git a/test/benchmark/lmbench/vfs_fstat_lat/bench_result.json b/test/benchmark/lmbench/vfs_fstat_lat/bench_result.json
deleted file mode 100644
index b3d13dd5..00000000
--- a/test/benchmark/lmbench/vfs_fstat_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Simple fstat: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[VFS] The cost of fstat",
-        "description": "lat_syscall fstat",
-        "unit": "\u00b5s",
-        "legend": "Average fstat latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/vfs_fstat_lat/bench_result.yaml b/test/benchmark/lmbench/vfs_fstat_lat/bench_result.yaml
new file mode 100644
index 00000000..1f4c1f6b
--- /dev/null
+++ b/test/benchmark/lmbench/vfs_fstat_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_syscall fstat
+  legend: Average fstat latency on {system}
+  title: '[VFS] The cost of fstat'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Simple fstat: +[0-9.]+'
diff --git a/test/benchmark/lmbench/vfs_open_lat/bench_result.json b/test/benchmark/lmbench/vfs_open_lat/bench_result.json
deleted file mode 100644
index c0fd0b5d..00000000
--- a/test/benchmark/lmbench/vfs_open_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Simple open\\/close: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[VFS] The cost of open+close",
-        "description": "lat_syscall open",
-        "unit": "\u00b5s",
-        "legend": "Average open latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/vfs_open_lat/bench_result.yaml b/test/benchmark/lmbench/vfs_open_lat/bench_result.yaml
new file mode 100644
index 00000000..1fc909c5
--- /dev/null
+++ b/test/benchmark/lmbench/vfs_open_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_syscall open
+  legend: Average open latency on {system}
+  title: '[VFS] The cost of open+close'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Simple open\/close: +[0-9.]+'
diff --git a/test/benchmark/lmbench/vfs_read_lat/bench_result.json b/test/benchmark/lmbench/vfs_read_lat/bench_result.json
deleted file mode 100644
index 2daab4d7..00000000
--- a/test/benchmark/lmbench/vfs_read_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "150%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Simple read: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[VFS] The cost of read (/dev/zero)",
-        "description": "lat_syscall read",
-        "unit": "\u00b5s",
-        "legend": "Average read latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/vfs_read_lat/bench_result.yaml b/test/benchmark/lmbench/vfs_read_lat/bench_result.yaml
new file mode 100644
index 00000000..4debeece
--- /dev/null
+++ b/test/benchmark/lmbench/vfs_read_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 150%
+chart:
+  description: lat_syscall read
+  legend: Average read latency on {system}
+  title: '[VFS] The cost of read (/dev/zero)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Simple read: +[0-9.]+'
diff --git a/test/benchmark/lmbench/vfs_read_pagecache_bw/bench_result.json b/test/benchmark/lmbench/vfs_read_pagecache_bw/bench_result.json
deleted file mode 100644
index 69ede061..00000000
--- a/test/benchmark/lmbench/vfs_read_pagecache_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "536.87 +[0-9.]+",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[VFS] The bandwidth of file reads via page cache",
-        "description": "bw_file_rd",
-        "unit": "MB/s",
-        "legend": "Average file read bandwidth on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/vfs_read_pagecache_bw/bench_result.yaml b/test/benchmark/lmbench/vfs_read_pagecache_bw/bench_result.yaml
new file mode 100644
index 00000000..96eefd37
--- /dev/null
+++ b/test/benchmark/lmbench/vfs_read_pagecache_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: bw_file_rd
+  legend: Average file read bandwidth on {system}
+  title: '[VFS] The bandwidth of file reads via page cache'
+  unit: MB/s
+result_extraction:
+  result_index: 2
+  search_pattern: 536.87 +[0-9.]+
diff --git a/test/benchmark/lmbench/vfs_select_lat/bench_result.json b/test/benchmark/lmbench/vfs_select_lat/bench_result.json
deleted file mode 100644
index ea026356..00000000
--- a/test/benchmark/lmbench/vfs_select_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Select on [0-9]+ fd's: +[0-9.]+",
-        "result_index": 5
-    },
-    "chart": {
-        "title": "[VFS] The cost of select (file fds)",
-        "description": "lat_select",
-        "unit": "\u00b5s",
-        "legend": "Average select file latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/vfs_select_lat/bench_result.yaml b/test/benchmark/lmbench/vfs_select_lat/bench_result.yaml
new file mode 100644
index 00000000..ff4b3911
--- /dev/null
+++ b/test/benchmark/lmbench/vfs_select_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_select
+  legend: Average select file latency on {system}
+  title: '[VFS] The cost of select (file fds)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 5
+  search_pattern: 'Select on [0-9]+ fd''s: +[0-9.]+'
diff --git a/test/benchmark/lmbench/vfs_stat_lat/bench_result.json b/test/benchmark/lmbench/vfs_stat_lat/bench_result.json
deleted file mode 100644
index 1404772d..00000000
--- a/test/benchmark/lmbench/vfs_stat_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Simple stat: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[VFS] The cost of stat",
-        "description": "lat_syscall stat",
-        "unit": "\u00b5s",
-        "legend": "Average stat latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/vfs_stat_lat/bench_result.yaml b/test/benchmark/lmbench/vfs_stat_lat/bench_result.yaml
new file mode 100644
index 00000000..362f7385
--- /dev/null
+++ b/test/benchmark/lmbench/vfs_stat_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: lat_syscall stat
+  legend: Average stat latency on {system}
+  title: '[VFS] The cost of stat'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Simple stat: +[0-9.]+'
diff --git a/test/benchmark/lmbench/vfs_write_lat/bench_result.json b/test/benchmark/lmbench/vfs_write_lat/bench_result.json
deleted file mode 100644
index 760b93b8..00000000
--- a/test/benchmark/lmbench/vfs_write_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "150%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Simple write: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[VFS] The cost of write (/dev/null)",
-        "description": "lat_syscall write",
-        "unit": "\u00b5s",
-        "legend": "Average write latency on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/lmbench/vfs_write_lat/bench_result.yaml b/test/benchmark/lmbench/vfs_write_lat/bench_result.yaml
new file mode 100644
index 00000000..53008a3b
--- /dev/null
+++ b/test/benchmark/lmbench/vfs_write_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 150%
+chart:
+  description: lat_syscall write
+  legend: Average write latency on {system}
+  title: '[VFS] The cost of write (/dev/null)'
+  unit: "\xB5s"   # Unicode for microsecond
+result_extraction:
+  result_index: 3
+  search_pattern: 'Simple write: +[0-9.]+'
diff --git a/test/benchmark/memcached/summary.json b/test/benchmark/memcached/summary.json
deleted file mode 100644
index 8c3bb7e6..00000000
--- a/test/benchmark/memcached/summary.json
+++ /dev/null
@@ -1,7 +0,0 @@
-{
-    "benchmarks": [
-        "t8_conc32_window10k",
-        "t8_conc32_window20k",
-        "t16_conc64_window10k"
-    ]
-}
\ No newline at end of file
diff --git a/test/benchmark/memcached/summary.yaml b/test/benchmark/memcached/summary.yaml
new file mode 100644
index 00000000..a841669e
--- /dev/null
+++ b/test/benchmark/memcached/summary.yaml
@@ -0,0 +1,4 @@
+benchmarks:
+  - t8_conc32_window10k
+  - t8_conc32_window20k
+  - t16_conc64_window10k
diff --git a/test/benchmark/memcached/t16_conc64_window10k/bench_result.json b/test/benchmark/memcached/t16_conc64_window10k/bench_result.json
deleted file mode 100644
index 2f73b705..00000000
--- a/test/benchmark/memcached/t16_conc64_window10k/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Run time: [0-9]+",
-        "result_index": 5
-    },
-    "chart": {
-        "title": "Memaslap 10% SET and 90% GET throughput (virtio-net, 16 client threads, 64 concurrency, 10k window)",
-        "description": "memaslap -t 30s -T 16 -c 64 -w 10k -S 1s",
-        "unit": "operations per second",
-        "legend": "Average OPS of SET and GET over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/memcached/t16_conc64_window10k/bench_result.yaml b/test/benchmark/memcached/t16_conc64_window10k/bench_result.yaml
new file mode 100644
index 00000000..1b649eda
--- /dev/null
+++ b/test/benchmark/memcached/t16_conc64_window10k/bench_result.yaml
@@ -0,0 +1,13 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: memaslap -t 30s -T 16 -c 64 -w 10k -S 1s
+  legend: Average OPS of SET and GET over virtio-net between Host Linux and Guest
+    {system}
+  title: Memaslap 10% SET and 90% GET throughput (virtio-net, 16 client threads, 64
+    concurrency, 10k window)
+  unit: operations per second
+result_extraction:
+  result_index: 5
+  search_pattern: 'Run time: [0-9]+'
diff --git a/test/benchmark/memcached/t8_conc32_window10k/bench_result.json b/test/benchmark/memcached/t8_conc32_window10k/bench_result.json
deleted file mode 100644
index 3998de6e..00000000
--- a/test/benchmark/memcached/t8_conc32_window10k/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Run time: [0-9]+",
-        "result_index": 5
-    },
-    "chart": {
-        "title": "Memaslap 10% SET and 90% GET throughput (virtio-net, 8 client threads, 32 concurrency, 10k window)",
-        "description": "memaslap -t 30s -T 8 -c 32 -w 10k -S 1s",
-        "unit": "operations per second",
-        "legend": "Average OPS of SET and GET over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/memcached/t8_conc32_window10k/bench_result.yaml b/test/benchmark/memcached/t8_conc32_window10k/bench_result.yaml
new file mode 100644
index 00000000..04cebf6f
--- /dev/null
+++ b/test/benchmark/memcached/t8_conc32_window10k/bench_result.yaml
@@ -0,0 +1,13 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: memaslap -t 30s -T 8 -c 32 -w 10k -S 1s
+  legend: Average OPS of SET and GET over virtio-net between Host Linux and Guest
+    {system}
+  title: Memaslap 10% SET and 90% GET throughput (virtio-net, 8 client threads, 32
+    concurrency, 10k window)
+  unit: operations per second
+result_extraction:
+  result_index: 5
+  search_pattern: 'Run time: [0-9]+'
diff --git a/test/benchmark/memcached/t8_conc32_window20k/bench_result.json b/test/benchmark/memcached/t8_conc32_window20k/bench_result.json
deleted file mode 100644
index 80f3a502..00000000
--- a/test/benchmark/memcached/t8_conc32_window20k/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Run time: [0-9]+",
-        "result_index": 5
-    },
-    "chart": {
-        "title": "Memaslap 10% SET and 90% GET throughput (virtio-net, 8 client threads, 32 concurrency, 20k window)",
-        "description": "memaslap -t 30s -T 8 -c 32 -w 20k -S 1s",
-        "unit": "operations per second",
-        "legend": "Average OPS of SET and GET over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/memcached/t8_conc32_window20k/bench_result.yaml b/test/benchmark/memcached/t8_conc32_window20k/bench_result.yaml
new file mode 100644
index 00000000..bab19877
--- /dev/null
+++ b/test/benchmark/memcached/t8_conc32_window20k/bench_result.yaml
@@ -0,0 +1,13 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: memaslap -t 30s -T 8 -c 32 -w 20k -S 1s
+  legend: Average OPS of SET and GET over virtio-net between Host Linux and Guest
+    {system}
+  title: Memaslap 10% SET and 90% GET throughput (virtio-net, 8 client threads, 32
+    concurrency, 20k window)
+  unit: operations per second
+result_extraction:
+  result_index: 5
+  search_pattern: 'Run time: [0-9]+'
diff --git a/test/benchmark/nginx/http_file16KB_bw/bench_result.json b/test/benchmark/nginx/http_file16KB_bw/bench_result.json
deleted file mode 100644
index dd66687c..00000000
--- a/test/benchmark/nginx/http_file16KB_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Requests per second: +[0-9.]+",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "Nginx HTTP request performance (virtio-net, 16KB file size)",
-        "description": "ab -n 10000 -c 1 http://10.0.2.15:8080/16384bytes.html",
-        "unit": "Requests per second",
-        "legend": "Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/nginx/http_file16KB_bw/bench_result.yaml b/test/benchmark/nginx/http_file16KB_bw/bench_result.yaml
new file mode 100644
index 00000000..488c4d1d
--- /dev/null
+++ b/test/benchmark/nginx/http_file16KB_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: ab -n 10000 -c 1 http://10.0.2.15:8080/16384bytes.html
+  legend: Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}
+  title: Nginx HTTP request performance (virtio-net, 16KB file size)
+  unit: Requests per second
+result_extraction:
+  result_index: 4
+  search_pattern: 'Requests per second: +[0-9.]+'
diff --git a/test/benchmark/nginx/http_file32KB_bw/bench_result.json b/test/benchmark/nginx/http_file32KB_bw/bench_result.json
deleted file mode 100644
index 33b133cb..00000000
--- a/test/benchmark/nginx/http_file32KB_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Requests per second: +[0-9.]+",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "Nginx HTTP request performance (virtio-net, 32KB file size)",
-        "description": "ab -n 10000 -c 1 http://10.0.2.15:8080/32768bytes.html",
-        "unit": "Requests per second",
-        "legend": "Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/nginx/http_file32KB_bw/bench_result.yaml b/test/benchmark/nginx/http_file32KB_bw/bench_result.yaml
new file mode 100644
index 00000000..14288414
--- /dev/null
+++ b/test/benchmark/nginx/http_file32KB_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: ab -n 10000 -c 1 http://10.0.2.15:8080/32768bytes.html
+  legend: Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}
+  title: Nginx HTTP request performance (virtio-net, 32KB file size)
+  unit: Requests per second
+result_extraction:
+  result_index: 4
+  search_pattern: 'Requests per second: +[0-9.]+'
diff --git a/test/benchmark/nginx/http_file4KB_bw/bench_result.json b/test/benchmark/nginx/http_file4KB_bw/bench_result.json
deleted file mode 100644
index 5ad6652e..00000000
--- a/test/benchmark/nginx/http_file4KB_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Requests per second: +[0-9.]+",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "Nginx HTTP request performance (virtio-net, 4KB file size)",
-        "description": "ab -n 10000 -c 1 http://10.0.2.15:8080/4096bytes.html",
-        "unit": "Requests per second",
-        "legend": "Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/nginx/http_file4KB_bw/bench_result.yaml b/test/benchmark/nginx/http_file4KB_bw/bench_result.yaml
new file mode 100644
index 00000000..1e6e3ce9
--- /dev/null
+++ b/test/benchmark/nginx/http_file4KB_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: ab -n 10000 -c 1 http://10.0.2.15:8080/4096bytes.html
+  legend: Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}
+  title: Nginx HTTP request performance (virtio-net, 4KB file size)
+  unit: Requests per second
+result_extraction:
+  result_index: 4
+  search_pattern: 'Requests per second: +[0-9.]+'
diff --git a/test/benchmark/nginx/http_file64KB_bw/bench_result.json b/test/benchmark/nginx/http_file64KB_bw/bench_result.json
deleted file mode 100644
index f1cb6392..00000000
--- a/test/benchmark/nginx/http_file64KB_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Requests per second: +[0-9.]+",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "Nginx HTTP request performance (virtio-net, 16KB file size)",
-        "description": "ab -n 10000 -c 1 http://10.0.2.15:8080/65536bytes.html",
-        "unit": "Requests per second",
-        "legend": "Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/nginx/http_file64KB_bw/bench_result.yaml b/test/benchmark/nginx/http_file64KB_bw/bench_result.yaml
new file mode 100644
index 00000000..49ff8ec9
--- /dev/null
+++ b/test/benchmark/nginx/http_file64KB_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: ab -n 10000 -c 1 http://10.0.2.15:8080/65536bytes.html
+  legend: Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}
+  title: Nginx HTTP request performance (virtio-net, 16KB file size)
+  unit: Requests per second
+result_extraction:
+  result_index: 4
+  search_pattern: 'Requests per second: +[0-9.]+'
diff --git a/test/benchmark/nginx/http_file8KB_bw/bench_result.json b/test/benchmark/nginx/http_file8KB_bw/bench_result.json
deleted file mode 100644
index 1dc50f13..00000000
--- a/test/benchmark/nginx/http_file8KB_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Requests per second: +[0-9.]+",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "Nginx HTTP request performance (virtio-net, 8KB file size)",
-        "description": "ab -n 10000 -c 1 http://10.0.2.15:8080/8192bytes.html",
-        "unit": "Requests per second",
-        "legend": "Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/nginx/http_file8KB_bw/bench_result.yaml b/test/benchmark/nginx/http_file8KB_bw/bench_result.yaml
new file mode 100644
index 00000000..0d9ef21d
--- /dev/null
+++ b/test/benchmark/nginx/http_file8KB_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: ab -n 10000 -c 1 http://10.0.2.15:8080/8192bytes.html
+  legend: Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}
+  title: Nginx HTTP request performance (virtio-net, 8KB file size)
+  unit: Requests per second
+result_extraction:
+  result_index: 4
+  search_pattern: 'Requests per second: +[0-9.]+'
diff --git a/test/benchmark/nginx/http_req10k_conc1_bw/bench_result.json b/test/benchmark/nginx/http_req10k_conc1_bw/bench_result.json
deleted file mode 100644
index 3730f518..00000000
--- a/test/benchmark/nginx/http_req10k_conc1_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Transfer rate: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "Nginx HTTP request performance with 1 concurrency and 10000 requests in total",
-        "description": "ab -n 10000 -c 1 http://localhost:8080/index.html",
-        "unit": "Kbytes/sec",
-        "legend": "Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/nginx/http_req10k_conc1_bw/bench_result.yaml b/test/benchmark/nginx/http_req10k_conc1_bw/bench_result.yaml
new file mode 100644
index 00000000..63574788
--- /dev/null
+++ b/test/benchmark/nginx/http_req10k_conc1_bw/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: ab -n 10000 -c 1 http://localhost:8080/index.html
+  legend: Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}
+  title: Nginx HTTP request performance with 1 concurrency and 10000 requests in total
+  unit: Kbytes/sec
+result_extraction:
+  result_index: 3
+  search_pattern: 'Transfer rate: +[0-9.]+'
diff --git a/test/benchmark/nginx/http_req10k_conc20_bw/bench_result.json b/test/benchmark/nginx/http_req10k_conc20_bw/bench_result.json
deleted file mode 100644
index 45310d16..00000000
--- a/test/benchmark/nginx/http_req10k_conc20_bw/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "Transfer rate: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "Nginx HTTP request performance with 20 concurrency and 10000 requests in total",
-        "description": "ab -n 10000 -c 20 http://localhost:8080/index.html",
-        "unit": "Kbytes/sec",
-        "legend": "Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/nginx/http_req10k_conc20_bw/bench_result.yaml b/test/benchmark/nginx/http_req10k_conc20_bw/bench_result.yaml
new file mode 100644
index 00000000..d5649297
--- /dev/null
+++ b/test/benchmark/nginx/http_req10k_conc20_bw/bench_result.yaml
@@ -0,0 +1,12 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: ab -n 10000 -c 20 http://localhost:8080/index.html
+  legend: Average HTTP Bandwidth over virtio-net between Host Linux and Guest {system}
+  title: Nginx HTTP request performance with 20 concurrency and 10000 requests in
+    total
+  unit: Kbytes/sec
+result_extraction:
+  result_index: 3
+  search_pattern: 'Transfer rate: +[0-9.]+'
diff --git a/test/benchmark/nginx/summary.json b/test/benchmark/nginx/summary.json
deleted file mode 100644
index 1e0b0252..00000000
--- a/test/benchmark/nginx/summary.json
+++ /dev/null
@@ -1,9 +0,0 @@
-{
-    "benchmarks": [
-        "http_file4KB_bw",
-        "http_file8KB_bw",
-        "http_file16KB_bw",
-        "http_file32KB_bw",
-        "http_file64KB_bw"
-    ]
-}
\ No newline at end of file
diff --git a/test/benchmark/nginx/summary.yaml b/test/benchmark/nginx/summary.yaml
new file mode 100644
index 00000000..fe58175c
--- /dev/null
+++ b/test/benchmark/nginx/summary.yaml
@@ -0,0 +1,8 @@
+benchmarks:
+  - http_file4KB_bw
+  - http_file8KB_bw
+  - http_file16KB_bw
+  - http_file32KB_bw
+  - http_file64KB_bw
+  # - http_req10k_conc1_bw
+  # - http_req10k_conc20_bw
diff --git a/test/benchmark/redis/get_100k_conc20_rps/bench_result.json b/test/benchmark/redis/get_100k_conc20_rps/bench_result.json
deleted file mode 100644
index 7945ddf0..00000000
--- a/test/benchmark/redis/get_100k_conc20_rps/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "throughput summary: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "Redis GET performance (virtio-net, 20 concurrency, 100K requests)",
-        "description": "redis-benchmark -n 100000 -c 20 -t get",
-        "unit": "request per second",
-        "legend": "Average RPS of GET over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/redis/get_100k_conc20_rps/bench_result.yaml b/test/benchmark/redis/get_100k_conc20_rps/bench_result.yaml
new file mode 100644
index 00000000..20d82cfa
--- /dev/null
+++ b/test/benchmark/redis/get_100k_conc20_rps/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: redis-benchmark -n 100000 -c 20 -t get
+  legend: Average RPS of GET over virtio-net between Host Linux and Guest {system}
+  title: Redis GET performance (virtio-net, 20 concurrency, 100K requests)
+  unit: request per second
+result_extraction:
+  result_index: 3
+  search_pattern: 'throughput summary: +[0-9.]+'
diff --git a/test/benchmark/redis/ping_inline_100k_conc20_rps/bench_result.json b/test/benchmark/redis/ping_inline_100k_conc20_rps/bench_result.json
deleted file mode 100644
index 761ff8e9..00000000
--- a/test/benchmark/redis/ping_inline_100k_conc20_rps/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "throughput summary: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "Redis PING INLINE performance (virtio-net, 20 concurrency, 100K requests)",
-        "description": "redis-benchmark -n 100000 -c 20 -t ping_inline",
-        "unit": "request per second",
-        "legend": "Average RPS of PING_INLINE over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/redis/ping_inline_100k_conc20_rps/bench_result.yaml b/test/benchmark/redis/ping_inline_100k_conc20_rps/bench_result.yaml
new file mode 100644
index 00000000..26771551
--- /dev/null
+++ b/test/benchmark/redis/ping_inline_100k_conc20_rps/bench_result.yaml
@@ -0,0 +1,12 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: redis-benchmark -n 100000 -c 20 -t ping_inline
+  legend: Average RPS of PING_INLINE over virtio-net between Host Linux and Guest
+    {system}
+  title: Redis PING INLINE performance (virtio-net, 20 concurrency, 100K requests)
+  unit: request per second
+result_extraction:
+  result_index: 3
+  search_pattern: 'throughput summary: +[0-9.]+'
diff --git a/test/benchmark/redis/ping_mbulk_100k_conc20_rps/bench_result.json b/test/benchmark/redis/ping_mbulk_100k_conc20_rps/bench_result.json
deleted file mode 100644
index 63287952..00000000
--- a/test/benchmark/redis/ping_mbulk_100k_conc20_rps/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "throughput summary: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "Redis PING MBULK performance (virtio-net, 20 concurrency, 100K requests)",
-        "description": "redis-benchmark -n 100000 -c 20 -t ping_mbulk",
-        "unit": "request per second",
-        "legend": "Average RPS of PING_MBULK over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/redis/ping_mbulk_100k_conc20_rps/bench_result.yaml b/test/benchmark/redis/ping_mbulk_100k_conc20_rps/bench_result.yaml
new file mode 100644
index 00000000..ea6ca3b0
--- /dev/null
+++ b/test/benchmark/redis/ping_mbulk_100k_conc20_rps/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: redis-benchmark -n 100000 -c 20 -t ping_mbulk
+  legend: Average RPS of PING_MBULK over virtio-net between Host Linux and Guest {system}
+  title: Redis PING MBULK performance (virtio-net, 20 concurrency, 100K requests)
+  unit: request per second
+result_extraction:
+  result_index: 3
+  search_pattern: 'throughput summary: +[0-9.]+'
diff --git a/test/benchmark/redis/set_100k_conc20_rps/bench_result.json b/test/benchmark/redis/set_100k_conc20_rps/bench_result.json
deleted file mode 100644
index ff2c4300..00000000
--- a/test/benchmark/redis/set_100k_conc20_rps/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "throughput summary: +[0-9.]+",
-        "result_index": 3
-    },
-    "chart": {
-        "title": "Redis SET performance (virtio-net, 20 concurrency, 100K requests)",
-        "description": "redis-benchmark -n 100000 -c 20 -t set",
-        "unit": "request per second",
-        "legend": "Average RPS of SET over virtio-net between Host Linux and Guest {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/redis/set_100k_conc20_rps/bench_result.yaml b/test/benchmark/redis/set_100k_conc20_rps/bench_result.yaml
new file mode 100644
index 00000000..6d614811
--- /dev/null
+++ b/test/benchmark/redis/set_100k_conc20_rps/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: true
+  threshold: 125%
+chart:
+  description: redis-benchmark -n 100000 -c 20 -t set
+  legend: Average RPS of SET over virtio-net between Host Linux and Guest {system}
+  title: Redis SET performance (virtio-net, 20 concurrency, 100K requests)
+  unit: request per second
+result_extraction:
+  result_index: 3
+  search_pattern: 'throughput summary: +[0-9.]+'
diff --git a/test/benchmark/redis/summary.json b/test/benchmark/redis/summary.json
deleted file mode 100644
index 3f5ee4f4..00000000
--- a/test/benchmark/redis/summary.json
+++ /dev/null
@@ -1,8 +0,0 @@
-{
-    "benchmarks": [
-        "ping_inline_100k_conc20_rps",
-        "ping_mbulk_100k_conc20_rps",
-        "get_100k_conc20_rps",
-        "set_100k_conc20_rps"
-    ]
-}
\ No newline at end of file
diff --git a/test/benchmark/redis/summary.yaml b/test/benchmark/redis/summary.yaml
new file mode 100644
index 00000000..24fc9d09
--- /dev/null
+++ b/test/benchmark/redis/summary.yaml
@@ -0,0 +1,5 @@
+benchmarks:
+  - ping_inline_100k_conc20_rps
+  - ping_mbulk_100k_conc20_rps
+  - get_100k_conc20_rps
+  - set_100k_conc20_rps
diff --git a/test/benchmark/schbench/smp1/bench_results/p50_rps.json b/test/benchmark/schbench/smp1/bench_results/p50_rps.json
deleted file mode 100644
index 074b108f..00000000
--- a/test/benchmark/schbench/smp1/bench_results/p50_rps.json
+++ /dev/null
@@ -1,17 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "\\*",
-        "nth_occurrence": 3,
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Scheduler] P50 RPS while max-loading the only CPU (SMP=1)",
-        "description": "schbench -F 256 -n 5 -r 90",
-        "unit": "requests per second",
-        "legend": "P50 RPS of schbench on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/schbench/smp1/bench_results/p50_rps.yaml b/test/benchmark/schbench/smp1/bench_results/p50_rps.yaml
new file mode 100644
index 00000000..17080eb4
--- /dev/null
+++ b/test/benchmark/schbench/smp1/bench_results/p50_rps.yaml
@@ -0,0 +1,12 @@
+alert:
+  bigger_is_better: true
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P50 RPS of schbench on {system}
+  title: '[Scheduler] P50 RPS while max-loading the only CPU (SMP=1)'
+  unit: requests per second
+result_extraction:
+  nth_occurrence: 3
+  result_index: 3
+  search_pattern: \*
diff --git a/test/benchmark/schbench/smp1/bench_results/p50_rps_smp1.yaml b/test/benchmark/schbench/smp1/bench_results/p50_rps_smp1.yaml
new file mode 100644
index 00000000..17080eb4
--- /dev/null
+++ b/test/benchmark/schbench/smp1/bench_results/p50_rps_smp1.yaml
@@ -0,0 +1,12 @@
+alert:
+  bigger_is_better: true
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P50 RPS of schbench on {system}
+  title: '[Scheduler] P50 RPS while max-loading the only CPU (SMP=1)'
+  unit: requests per second
+result_extraction:
+  nth_occurrence: 3
+  result_index: 3
+  search_pattern: \*
diff --git a/test/benchmark/schbench/smp1/bench_results/p99_request_latency.json b/test/benchmark/schbench/smp1/bench_results/p99_request_latency.json
deleted file mode 100644
index c4b5c7b0..00000000
--- a/test/benchmark/schbench/smp1/bench_results/p99_request_latency.json
+++ /dev/null
@@ -1,17 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "\\*",
-        "nth_occurrence": 2,
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Scheduler] P99 request latency while max-loading the only CPU (SMP=1)",
-        "description": "schbench -F 256 -n 5 -r 90",
-        "unit": "\u00b5s",
-        "legend": "P99 request latency of schbench on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/schbench/smp1/bench_results/p99_request_latency.yaml b/test/benchmark/schbench/smp1/bench_results/p99_request_latency.yaml
new file mode 100644
index 00000000..1e5cdbec
--- /dev/null
+++ b/test/benchmark/schbench/smp1/bench_results/p99_request_latency.yaml
@@ -0,0 +1,12 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P99 request latency of schbench on {system}
+  title: '[Scheduler] P99 request latency while max-loading the only CPU (SMP=1)'
+  unit: "\xB5s"
+result_extraction:
+  nth_occurrence: 2
+  result_index: 3
+  search_pattern: \*
diff --git a/test/benchmark/schbench/smp1/bench_results/p99_request_latency_smp1.yaml b/test/benchmark/schbench/smp1/bench_results/p99_request_latency_smp1.yaml
new file mode 100644
index 00000000..1e5cdbec
--- /dev/null
+++ b/test/benchmark/schbench/smp1/bench_results/p99_request_latency_smp1.yaml
@@ -0,0 +1,12 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P99 request latency of schbench on {system}
+  title: '[Scheduler] P99 request latency while max-loading the only CPU (SMP=1)'
+  unit: "\xB5s"
+result_extraction:
+  nth_occurrence: 2
+  result_index: 3
+  search_pattern: \*
diff --git a/test/benchmark/schbench/smp1/bench_results/p99_wakeup_latency.json b/test/benchmark/schbench/smp1/bench_results/p99_wakeup_latency.json
deleted file mode 100644
index 0fabc152..00000000
--- a/test/benchmark/schbench/smp1/bench_results/p99_wakeup_latency.json
+++ /dev/null
@@ -1,17 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "\\*",
-        "nth_occurrence": 1,
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Scheduler] P99 wakeup latency while max-loading the only CPU (SMP=1)",
-        "description": "schbench -F 256 -n 5 -r 90",
-        "unit": "\u00b5s",
-        "legend": "P99 wakeup latency of schbench on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/schbench/smp1/bench_results/p99_wakeup_latency.yaml b/test/benchmark/schbench/smp1/bench_results/p99_wakeup_latency.yaml
new file mode 100644
index 00000000..0564166b
--- /dev/null
+++ b/test/benchmark/schbench/smp1/bench_results/p99_wakeup_latency.yaml
@@ -0,0 +1,12 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P99 wakeup latency of schbench on {system}
+  title: '[Scheduler] P99 wakeup latency while max-loading the only CPU (SMP=1)'
+  unit: "\xB5s"
+result_extraction:
+  nth_occurrence: 1
+  result_index: 3
+  search_pattern: \*
diff --git a/test/benchmark/schbench/smp1/bench_results/p99_wakeup_latency_smp1.yaml b/test/benchmark/schbench/smp1/bench_results/p99_wakeup_latency_smp1.yaml
new file mode 100644
index 00000000..0564166b
--- /dev/null
+++ b/test/benchmark/schbench/smp1/bench_results/p99_wakeup_latency_smp1.yaml
@@ -0,0 +1,12 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P99 wakeup latency of schbench on {system}
+  title: '[Scheduler] P99 wakeup latency while max-loading the only CPU (SMP=1)'
+  unit: "\xB5s"
+result_extraction:
+  nth_occurrence: 1
+  result_index: 3
+  search_pattern: \*
diff --git a/test/benchmark/schbench/smp8/bench_results/p50_rps.json b/test/benchmark/schbench/smp8/bench_results/p50_rps.json
deleted file mode 100644
index 19305d13..00000000
--- a/test/benchmark/schbench/smp8/bench_results/p50_rps.json
+++ /dev/null
@@ -1,20 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": true
-    },
-    "result_extraction": {
-        "search_pattern": "\\*",
-        "nth_occurrence": 3,
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Scheduler] P50 RPS while max-loading the only CPU (SMP=8)",
-        "description": "schbench -F 256 -n 5 -r 90",
-        "unit": "requests per second",
-        "legend": "P50 RPS of schbench on {system}"
-    },
-    "runtime_config": {
-        "smp": 8
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/schbench/smp8/bench_results/p50_rps.yaml b/test/benchmark/schbench/smp8/bench_results/p50_rps.yaml
new file mode 100644
index 00000000..83e16849
--- /dev/null
+++ b/test/benchmark/schbench/smp8/bench_results/p50_rps.yaml
@@ -0,0 +1,14 @@
+alert:
+  bigger_is_better: true
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P50 RPS of schbench on {system}
+  title: '[Scheduler] P50 RPS while max-loading the only CPU (SMP=8)'
+  unit: requests per second
+result_extraction:
+  nth_occurrence: 3
+  result_index: 3
+  search_pattern: \*
+runtime_config:
+  smp: 8
diff --git a/test/benchmark/schbench/smp8/bench_results/p50_rps_smp8.yaml b/test/benchmark/schbench/smp8/bench_results/p50_rps_smp8.yaml
new file mode 100644
index 00000000..83e16849
--- /dev/null
+++ b/test/benchmark/schbench/smp8/bench_results/p50_rps_smp8.yaml
@@ -0,0 +1,14 @@
+alert:
+  bigger_is_better: true
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P50 RPS of schbench on {system}
+  title: '[Scheduler] P50 RPS while max-loading the only CPU (SMP=8)'
+  unit: requests per second
+result_extraction:
+  nth_occurrence: 3
+  result_index: 3
+  search_pattern: \*
+runtime_config:
+  smp: 8
diff --git a/test/benchmark/schbench/smp8/bench_results/p99_request_latency.json b/test/benchmark/schbench/smp8/bench_results/p99_request_latency.json
deleted file mode 100644
index 3088310c..00000000
--- a/test/benchmark/schbench/smp8/bench_results/p99_request_latency.json
+++ /dev/null
@@ -1,20 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "\\*",
-        "nth_occurrence": 2,
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Scheduler] P99 request latency while max-loading the only CPU (SMP=8)",
-        "description": "schbench -F 256 -n 5 -r 90",
-        "unit": "\u00b5s",
-        "legend": "P99 request latency of schbench on {system}"
-    },
-    "runtime_config": {
-        "smp": 8
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/schbench/smp8/bench_results/p99_request_latency.yaml b/test/benchmark/schbench/smp8/bench_results/p99_request_latency.yaml
new file mode 100644
index 00000000..829a6d3a
--- /dev/null
+++ b/test/benchmark/schbench/smp8/bench_results/p99_request_latency.yaml
@@ -0,0 +1,14 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P99 request latency of schbench on {system}
+  title: '[Scheduler] P99 request latency while max-loading the only CPU (SMP=8)'
+  unit: "\xB5s"
+result_extraction:
+  nth_occurrence: 2
+  result_index: 3
+  search_pattern: \*
+runtime_config:
+  smp: 8
diff --git a/test/benchmark/schbench/smp8/bench_results/p99_request_latency_smp8.yaml b/test/benchmark/schbench/smp8/bench_results/p99_request_latency_smp8.yaml
new file mode 100644
index 00000000..829a6d3a
--- /dev/null
+++ b/test/benchmark/schbench/smp8/bench_results/p99_request_latency_smp8.yaml
@@ -0,0 +1,14 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P99 request latency of schbench on {system}
+  title: '[Scheduler] P99 request latency while max-loading the only CPU (SMP=8)'
+  unit: "\xB5s"
+result_extraction:
+  nth_occurrence: 2
+  result_index: 3
+  search_pattern: \*
+runtime_config:
+  smp: 8
diff --git a/test/benchmark/schbench/smp8/bench_results/p99_wakeup_latency.json b/test/benchmark/schbench/smp8/bench_results/p99_wakeup_latency.json
deleted file mode 100644
index 210a6baf..00000000
--- a/test/benchmark/schbench/smp8/bench_results/p99_wakeup_latency.json
+++ /dev/null
@@ -1,20 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "\\*",
-        "nth_occurrence": 1,
-        "result_index": 3
-    },
-    "chart": {
-        "title": "[Scheduler] P99 wakeup latency while max-loading the only CPU (SMP=8)",
-        "description": "schbench -F 256 -n 5 -r 90",
-        "unit": "\u00b5s",
-        "legend": "P99 wakeup latency of schbench on {system}"
-    },
-    "runtime_config": {
-        "smp": 8
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/schbench/smp8/bench_results/p99_wakeup_latency.yaml b/test/benchmark/schbench/smp8/bench_results/p99_wakeup_latency.yaml
new file mode 100644
index 00000000..af3e2e31
--- /dev/null
+++ b/test/benchmark/schbench/smp8/bench_results/p99_wakeup_latency.yaml
@@ -0,0 +1,14 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P99 wakeup latency of schbench on {system}
+  title: '[Scheduler] P99 wakeup latency while max-loading the only CPU (SMP=8)'
+  unit: "\xB5s"
+result_extraction:
+  nth_occurrence: 1
+  result_index: 3
+  search_pattern: \*
+runtime_config:
+  smp: 8
diff --git a/test/benchmark/schbench/smp8/bench_results/p99_wakeup_latency_smp8.yaml b/test/benchmark/schbench/smp8/bench_results/p99_wakeup_latency_smp8.yaml
new file mode 100644
index 00000000..af3e2e31
--- /dev/null
+++ b/test/benchmark/schbench/smp8/bench_results/p99_wakeup_latency_smp8.yaml
@@ -0,0 +1,14 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: schbench -F 256 -n 5 -r 90
+  legend: P99 wakeup latency of schbench on {system}
+  title: '[Scheduler] P99 wakeup latency while max-loading the only CPU (SMP=8)'
+  unit: "\xB5s"
+result_extraction:
+  nth_occurrence: 1
+  result_index: 3
+  search_pattern: \*
+runtime_config:
+  smp: 8
diff --git a/test/benchmark/schbench/summary.json b/test/benchmark/schbench/summary.json
deleted file mode 100644
index 335faa68..00000000
--- a/test/benchmark/schbench/summary.json
+++ /dev/null
@@ -1,4 +0,0 @@
-{
-    "benchmarks": [
-    ]
-}
\ No newline at end of file
diff --git a/test/benchmark/schbench/summary.yaml b/test/benchmark/schbench/summary.yaml
new file mode 100644
index 00000000..6953ccbd
--- /dev/null
+++ b/test/benchmark/schbench/summary.yaml
@@ -0,0 +1 @@
+benchmarks: []
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_between.json b/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_between.json
deleted file mode 100644
index 6022fafc..00000000
--- a/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_between.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ DELETEs, numeric BETWEEN, indexed....",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ext2] The deletes-between result of sqlite speedtest",
-        "description": "sqlite-speed-test: 10000 DELETEs, numeric BETWEEN, indexed",
-        "unit": "second",
-        "legend": "The speed of deletes-between on {system} (Ext2)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_between.yaml b/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_between.yaml
new file mode 100644
index 00000000..d79c6dad
--- /dev/null
+++ b/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_between.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 10000 DELETEs, numeric BETWEEN, indexed'
+  legend: The speed of deletes-between on {system} (Ext2)
+  title: '[Ext2] The deletes-between result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 8
+  search_pattern: '[0-9]+ DELETEs, numeric BETWEEN, indexed....'
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_individual.json b/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_individual.json
deleted file mode 100644
index 6b35841c..00000000
--- a/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_individual.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ DELETEs of individual rows....",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ext2] The deletes-individual result of sqlite speedtest",
-        "description": "sqlite-speed-test: 50000 DELETEs of individual rows",
-        "unit": "second",
-        "legend": "The speed of deletes-individual on {system} (Ext2)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_individual.yaml b/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_individual.yaml
new file mode 100644
index 00000000..64248bd6
--- /dev/null
+++ b/test/benchmark/sqlite/ext2_benchmarks/bench_results/deletes_individual.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 50000 DELETEs of individual rows'
+  legend: The speed of deletes-individual on {system} (Ext2)
+  title: '[Ext2] The deletes-individual result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 8
+  search_pattern: '[0-9]+ DELETEs of individual rows....'
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/refill_replace.json b/test/benchmark/sqlite/ext2_benchmarks/bench_results/refill_replace.json
deleted file mode 100644
index cf97b260..00000000
--- a/test/benchmark/sqlite/ext2_benchmarks/bench_results/refill_replace.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Refill two [0-9]+-row tables using REPLACE....",
-        "result_index": 9
-    },
-    "chart": {
-        "title": "[Ext2] The refill-replace result of sqlite speedtest",
-        "description": "sqlite-speed-test: Refill two 50000-row tables using REPLACE",
-        "unit": "second",
-        "legend": "The speed of refill-replace on {system} (Ext2)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/refill_replace.yaml b/test/benchmark/sqlite/ext2_benchmarks/bench_results/refill_replace.yaml
new file mode 100644
index 00000000..3ad13e9e
--- /dev/null
+++ b/test/benchmark/sqlite/ext2_benchmarks/bench_results/refill_replace.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: Refill two 50000-row tables using REPLACE'
+  legend: The speed of refill-replace on {system} (Ext2)
+  title: '[Ext2] The refill-replace result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 9
+  search_pattern: Refill two [0-9]+-row tables using REPLACE....
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_ipk.json b/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_ipk.json
deleted file mode 100644
index 02b759c1..00000000
--- a/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_ipk.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ SELECTS on an IPK....",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ext2] The selects-ipk result of sqlite speedtest",
-        "description": "sqlite-speed-test: 70000 SELECTS on an IPK",
-        "unit": "second",
-        "legend": "The speed of selects-ipk on {system} (Ext2)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_ipk.yaml b/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_ipk.yaml
new file mode 100644
index 00000000..0478af06
--- /dev/null
+++ b/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_ipk.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 70000 SELECTS on an IPK'
+  legend: The speed of selects-ipk on {system} (Ext2)
+  title: '[Ext2] The selects-ipk result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 8
+  search_pattern: '[0-9]+ SELECTS on an IPK....'
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_text_pk.json b/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_text_pk.json
deleted file mode 100644
index 6f26bc0a..00000000
--- a/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_text_pk.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ SELECTS on a TEXT PK....",
-        "result_index": 9
-    },
-    "chart": {
-        "title": "[Ext2] The selects-text-pk result of sqlite speedtest",
-        "description": "sqlite-speed-test: 70000 SELECTS on a TEXT PK",
-        "unit": "second",
-        "legend": "The speed of selects-text-pk on {system} (Ext2)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_text_pk.yaml b/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_text_pk.yaml
new file mode 100644
index 00000000..a4305d47
--- /dev/null
+++ b/test/benchmark/sqlite/ext2_benchmarks/bench_results/selects_text_pk.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 70000 SELECTS on a TEXT PK'
+  legend: The speed of selects-text-pk on {system} (Ext2)
+  title: '[Ext2] The selects-text-pk result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 9
+  search_pattern: '[0-9]+ SELECTS on a TEXT PK....'
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/total.json b/test/benchmark/sqlite/ext2_benchmarks/bench_results/total.json
deleted file mode 100644
index e6561f42..00000000
--- a/test/benchmark/sqlite/ext2_benchmarks/bench_results/total.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "TOTAL....",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Ext2] The total result of sqlite speedtest",
-        "description": "sqlite-speed-test: total",
-        "unit": "second",
-        "legend": "Total speed on {system} (Ext2)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/total.yaml b/test/benchmark/sqlite/ext2_benchmarks/bench_results/total.yaml
new file mode 100644
index 00000000..40cb1e70
--- /dev/null
+++ b/test/benchmark/sqlite/ext2_benchmarks/bench_results/total.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: total'
+  legend: Total speed on {system} (Ext2)
+  title: '[Ext2] The total result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 2
+  search_pattern: TOTAL....
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_between.json b/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_between.json
deleted file mode 100644
index 0355fc6a..00000000
--- a/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_between.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ UPDATES, numeric BETWEEN, indexed....",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ext2] The updates-between result of sqlite speedtest",
-        "description": "sqlite-speed-test: 10000 UPDATES, numeric BETWEEN, indexed",
-        "unit": "second",
-        "legend": "The speed of updates-between on {system} (Ext2)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_between.yaml b/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_between.yaml
new file mode 100644
index 00000000..39f9b763
--- /dev/null
+++ b/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_between.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 10000 UPDATES, numeric BETWEEN, indexed'
+  legend: The speed of updates-between on {system} (Ext2)
+  title: '[Ext2] The updates-between result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 8
+  search_pattern: '[0-9]+ UPDATES, numeric BETWEEN, indexed....'
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_big_one.json b/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_big_one.json
deleted file mode 100644
index a56c6646..00000000
--- a/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_big_one.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "One big UPDATE of the whole [0-9]+-row table....",
-        "result_index": 11
-    },
-    "chart": {
-        "title": "[Ext2] The updates-big-one result of sqlite speedtest",
-        "description": "sqlite-speed-test: One big UPDATE of the whole 50000-row table",
-        "unit": "second",
-        "legend": "The speed of updates-big-one on {system} (Ext2)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_big_one.yaml b/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_big_one.yaml
new file mode 100644
index 00000000..f0349c04
--- /dev/null
+++ b/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_big_one.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: One big UPDATE of the whole 50000-row table'
+  legend: The speed of updates-big-one on {system} (Ext2)
+  title: '[Ext2] The updates-big-one result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 11
+  search_pattern: One big UPDATE of the whole [0-9]+-row table....
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_individual.json b/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_individual.json
deleted file mode 100644
index 9f6976bd..00000000
--- a/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_individual.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ UPDATES of individual rows....",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ext2] The updates-individual result of sqlite speedtest",
-        "description": "sqlite-speed-test: 50000 UPDATES of individual rows",
-        "unit": "second",
-        "legend": "The speed of updates-individual on {system} (Ext2)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_individual.yaml b/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_individual.yaml
new file mode 100644
index 00000000..6bce7a02
--- /dev/null
+++ b/test/benchmark/sqlite/ext2_benchmarks/bench_results/updates_individual.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 50000 UPDATES of individual rows'
+  legend: The speed of updates-individual on {system} (Ext2)
+  title: '[Ext2] The updates-individual result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 8
+  search_pattern: '[0-9]+ UPDATES of individual rows....'
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/vacuum.json b/test/benchmark/sqlite/ext2_benchmarks/bench_results/vacuum.json
deleted file mode 100644
index 3531a1ac..00000000
--- a/test/benchmark/sqlite/ext2_benchmarks/bench_results/vacuum.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "VACUUM....",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "[Ext2] The 'VACUUM' result of sqlite speedtest",
-        "description": "sqlite-speed-test: repacking database file into a minimal amount of disk space",
-        "unit": "second",
-        "legend": "The speed of VACUUM on {system} (Ext2)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ext2_benchmarks/bench_results/vacuum.yaml b/test/benchmark/sqlite/ext2_benchmarks/bench_results/vacuum.yaml
new file mode 100644
index 00000000..debab04a
--- /dev/null
+++ b/test/benchmark/sqlite/ext2_benchmarks/bench_results/vacuum.yaml
@@ -0,0 +1,12 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: repacking database file into a minimal amount of
+    disk space'
+  legend: The speed of VACUUM on {system} (Ext2)
+  title: '[Ext2] The ''VACUUM'' result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 4
+  search_pattern: VACUUM....
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_between.json b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_between.json
deleted file mode 100644
index 9c6bf2b6..00000000
--- a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_between.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ DELETEs, numeric BETWEEN, indexed....",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ramfs] The deletes-between result of sqlite speedtest",
-        "description": "sqlite-speed-test: 10000 DELETEs, numeric BETWEEN, indexed",
-        "unit": "second",
-        "legend": "The speed of deletes-between on {system} (Ramfs)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_between.yaml b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_between.yaml
new file mode 100644
index 00000000..62f1fbad
--- /dev/null
+++ b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_between.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 10000 DELETEs, numeric BETWEEN, indexed'
+  legend: The speed of deletes-between on {system} (Ramfs)
+  title: '[Ramfs] The deletes-between result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 8
+  search_pattern: '[0-9]+ DELETEs, numeric BETWEEN, indexed....'
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_individual.json b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_individual.json
deleted file mode 100644
index 628b48b8..00000000
--- a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_individual.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ DELETEs of individual rows....",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ramfs] The deletes-individual result of sqlite speedtest",
-        "description": "sqlite-speed-test: 50000 DELETEs of individual rows",
-        "unit": "second",
-        "legend": "The speed of deletes-individual on {system} (Ramfs)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_individual.yaml b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_individual.yaml
new file mode 100644
index 00000000..802793bf
--- /dev/null
+++ b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/deletes_individual.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 50000 DELETEs of individual rows'
+  legend: The speed of deletes-individual on {system} (Ramfs)
+  title: '[Ramfs] The deletes-individual result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 8
+  search_pattern: '[0-9]+ DELETEs of individual rows....'
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/refill_replace.json b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/refill_replace.json
deleted file mode 100644
index 77d159f8..00000000
--- a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/refill_replace.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "Refill two [0-9]+-row tables using REPLACE....",
-        "result_index": 9
-    },
-    "chart": {
-        "title": "[Ramfs] The refill-replace result of sqlite speedtest",
-        "description": "sqlite-speed-test: Refill two 50000-row tables using REPLACE",
-        "unit": "second",
-        "legend": "The speed of refill-replace on {system} (Ramfs)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/refill_replace.yaml b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/refill_replace.yaml
new file mode 100644
index 00000000..6c7b01bb
--- /dev/null
+++ b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/refill_replace.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: Refill two 50000-row tables using REPLACE'
+  legend: The speed of refill-replace on {system} (Ramfs)
+  title: '[Ramfs] The refill-replace result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 9
+  search_pattern: Refill two [0-9]+-row tables using REPLACE....
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_ipk.json b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_ipk.json
deleted file mode 100644
index fbd871dd..00000000
--- a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_ipk.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ SELECTS on an IPK....",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ramfs] The selects-ipk result of sqlite speedtest",
-        "description": "sqlite-speed-test: 70000 SELECTS on an IPK",
-        "unit": "second",
-        "legend": "The speed of selects-ipk on {system} (Ramfs)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_ipk.yaml b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_ipk.yaml
new file mode 100644
index 00000000..f1d0d77c
--- /dev/null
+++ b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_ipk.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 70000 SELECTS on an IPK'
+  legend: The speed of selects-ipk on {system} (Ramfs)
+  title: '[Ramfs] The selects-ipk result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 8
+  search_pattern: '[0-9]+ SELECTS on an IPK....'
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_text_pk.json b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_text_pk.json
deleted file mode 100644
index 397115a8..00000000
--- a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_text_pk.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ SELECTS on a TEXT PK....",
-        "result_index": 9
-    },
-    "chart": {
-        "title": "[Ramfs] The selects-text-pk result of sqlite speedtest",
-        "description": "sqlite-speed-test: 70000 SELECTS on a TEXT PK",
-        "unit": "second",
-        "legend": "The speed of selects-text-pk on {system} (Ramfs)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_text_pk.yaml b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_text_pk.yaml
new file mode 100644
index 00000000..4110a321
--- /dev/null
+++ b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/selects_text_pk.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 70000 SELECTS on a TEXT PK'
+  legend: The speed of selects-text-pk on {system} (Ramfs)
+  title: '[Ramfs] The selects-text-pk result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 9
+  search_pattern: '[0-9]+ SELECTS on a TEXT PK....'
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/total.json b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/total.json
deleted file mode 100644
index aa0e06a3..00000000
--- a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/total.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "TOTAL....",
-        "result_index": 2
-    },
-    "chart": {
-        "title": "[Ramfs] The total result of sqlite speedtest",
-        "description": "sqlite-speed-test: total",
-        "unit": "second",
-        "legend": "Total speed on {system} (Ramfs)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/total.yaml b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/total.yaml
new file mode 100644
index 00000000..1a7466d9
--- /dev/null
+++ b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/total.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: total'
+  legend: Total speed on {system} (Ramfs)
+  title: '[Ramfs] The total result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 2
+  search_pattern: TOTAL....
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_between.json b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_between.json
deleted file mode 100644
index 2a410113..00000000
--- a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_between.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ UPDATES, numeric BETWEEN, indexed....",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ramfs] The updates-between result of sqlite speedtest",
-        "description": "sqlite-speed-test: 10000 UPDATES, numeric BETWEEN, indexed",
-        "unit": "second",
-        "legend": "The speed of updates-between on {system} (Ramfs)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_between.yaml b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_between.yaml
new file mode 100644
index 00000000..3447aebf
--- /dev/null
+++ b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_between.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 10000 UPDATES, numeric BETWEEN, indexed'
+  legend: The speed of updates-between on {system} (Ramfs)
+  title: '[Ramfs] The updates-between result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 8
+  search_pattern: '[0-9]+ UPDATES, numeric BETWEEN, indexed....'
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_big_one.json b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_big_one.json
deleted file mode 100644
index 4e3f6d51..00000000
--- a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_big_one.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "One big UPDATE of the whole [0-9]+-row table....",
-        "result_index": 11
-    },
-    "chart": {
-        "title": "[Ramfs] The updates-big-one result of sqlite speedtest",
-        "description": "sqlite-speed-test: One big UPDATE of the whole 50000-row table",
-        "unit": "second",
-        "legend": "The speed of updates-big-one on {system} (Ramfs)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_big_one.yaml b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_big_one.yaml
new file mode 100644
index 00000000..2031efcc
--- /dev/null
+++ b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_big_one.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: One big UPDATE of the whole 50000-row table'
+  legend: The speed of updates-big-one on {system} (Ramfs)
+  title: '[Ramfs] The updates-big-one result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 11
+  search_pattern: One big UPDATE of the whole [0-9]+-row table....
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_individual.json b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_individual.json
deleted file mode 100644
index 6a2d9b3e..00000000
--- a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_individual.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "[0-9]+ UPDATES of individual rows....",
-        "result_index": 8
-    },
-    "chart": {
-        "title": "[Ramfs] The updates-individual result of sqlite speedtest",
-        "description": "sqlite-speed-test: 50000 UPDATES of individual rows",
-        "unit": "second",
-        "legend": "The speed of updates-individual on {system} (Ramfs)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_individual.yaml b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_individual.yaml
new file mode 100644
index 00000000..f3c065e8
--- /dev/null
+++ b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/updates_individual.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: 50000 UPDATES of individual rows'
+  legend: The speed of updates-individual on {system} (Ramfs)
+  title: '[Ramfs] The updates-individual result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 8
+  search_pattern: '[0-9]+ UPDATES of individual rows....'
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/vacuum.json b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/vacuum.json
deleted file mode 100644
index f81cc439..00000000
--- a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/vacuum.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "125%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "VACUUM....",
-        "result_index": 4
-    },
-    "chart": {
-        "title": "[Ramfs] The 'VACUUM' result of sqlite speedtest",
-        "description": "sqlite-speed-test: repacking database file into a minimal amount of disk space",
-        "unit": "second",
-        "legend": "The speed of VACUUM on {system} (Ramfs)"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/ramfs_benchmarks/bench_results/vacuum.yaml b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/vacuum.yaml
new file mode 100644
index 00000000..ca64c0f2
--- /dev/null
+++ b/test/benchmark/sqlite/ramfs_benchmarks/bench_results/vacuum.yaml
@@ -0,0 +1,12 @@
+alert:
+  bigger_is_better: false
+  threshold: 125%
+chart:
+  description: 'sqlite-speed-test: repacking database file into a minimal amount of
+    disk space'
+  legend: The speed of VACUUM on {system} (Ramfs)
+  title: '[Ramfs] The ''VACUUM'' result of sqlite speedtest'
+  unit: second
+result_extraction:
+  result_index: 4
+  search_pattern: VACUUM....
diff --git a/test/benchmark/sqlite/summary.json b/test/benchmark/sqlite/summary.json
deleted file mode 100644
index 2d036645..00000000
--- a/test/benchmark/sqlite/summary.json
+++ /dev/null
@@ -1,24 +0,0 @@
-{
-    "benchmarks": [
-        "ext2_deletes_between",
-        "ext2_deletes_individual",
-        "ext2_refill_replace",
-        "ext2_selects_ipk",
-        "ext2_selects_text_pk",
-        "ext2_updates_between",
-        "ext2_updates_individual",
-        "ext2_updates_big_one",
-        "ext2_vacuum",
-        "ext2_total",
-        "ramfs_deletes_between",
-        "ramfs_deletes_individual",
-        "ramfs_refill_replace",
-        "ramfs_selects_ipk",
-        "ramfs_selects_text_pk",
-        "ramfs_updates_between",
-        "ramfs_updates_individual",
-        "ramfs_updates_big_one",
-        "ramfs_vacuum",
-        "ramfs_total"
-    ]
-}
\ No newline at end of file
diff --git a/test/benchmark/sqlite/summary.yaml b/test/benchmark/sqlite/summary.yaml
new file mode 100644
index 00000000..b37c7829
--- /dev/null
+++ b/test/benchmark/sqlite/summary.yaml
@@ -0,0 +1,21 @@
+benchmarks:
+  - ext2_deletes_between
+  - ext2_deletes_individual
+  - ext2_refill_replace
+  - ext2_selects_ipk
+  - ext2_selects_text_pk
+  - ext2_updates_between
+  - ext2_updates_individual
+  - ext2_updates_big_one
+  - ext2_vacuum
+  - ext2_total
+  - ramfs_deletes_between
+  - ramfs_deletes_individual
+  - ramfs_refill_replace
+  - ramfs_selects_ipk
+  - ramfs_selects_text_pk
+  - ramfs_updates_between
+  - ramfs_updates_individual
+  - ramfs_updates_big_one
+  - ramfs_vacuum
+  - ramfs_total
diff --git a/test/benchmark/sysbench/cpu_lat/bench_result.json b/test/benchmark/sysbench/cpu_lat/bench_result.json
deleted file mode 100644
index 8baacaa9..00000000
--- a/test/benchmark/sysbench/cpu_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "avg: +[0-9.]+",
-        "result_index": "NF"
-    },
-    "chart": {
-        "title": "[CPU] CPU performance",
-        "description": "sysbench cpu",
-        "unit": "ms",
-        "legend": "Average Execution Time per CPU on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sysbench/cpu_lat/bench_result.yaml b/test/benchmark/sysbench/cpu_lat/bench_result.yaml
new file mode 100644
index 00000000..24c00fed
--- /dev/null
+++ b/test/benchmark/sysbench/cpu_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: sysbench cpu
+  legend: Average Execution Time per CPU on {system}
+  title: '[CPU] CPU performance'
+  unit: ms
+result_extraction:
+  result_index: NF
+  search_pattern: 'avg: +[0-9.]+'
diff --git a/test/benchmark/sysbench/summary.json b/test/benchmark/sysbench/summary.json
deleted file mode 100644
index 9c5b805e..00000000
--- a/test/benchmark/sysbench/summary.json
+++ /dev/null
@@ -1,6 +0,0 @@
-{
-    "benchmarks": [
-        "cpu_lat",
-        "thread_lat"
-    ]
-}
\ No newline at end of file
diff --git a/test/benchmark/sysbench/summary.yaml b/test/benchmark/sysbench/summary.yaml
new file mode 100644
index 00000000..f82048b6
--- /dev/null
+++ b/test/benchmark/sysbench/summary.yaml
@@ -0,0 +1,3 @@
+benchmarks:
+- cpu_lat
+- thread_lat
diff --git a/test/benchmark/sysbench/thread_lat/bench_result.json b/test/benchmark/sysbench/thread_lat/bench_result.json
deleted file mode 100644
index dfb7b09c..00000000
--- a/test/benchmark/sysbench/thread_lat/bench_result.json
+++ /dev/null
@@ -1,16 +0,0 @@
-{
-    "alert": {
-        "threshold": "130%",
-        "bigger_is_better": false
-    },
-    "result_extraction": {
-        "search_pattern": "avg: +[0-9.]+",
-        "result_index": "NF"
-    },
-    "chart": {
-        "title": "[Threads] Threads performance",
-        "description": "sysbench threads",
-        "unit": "ms",
-        "legend": "Average Execution Time per Thread on {system}"
-    }
-}
\ No newline at end of file
diff --git a/test/benchmark/sysbench/thread_lat/bench_result.yaml b/test/benchmark/sysbench/thread_lat/bench_result.yaml
new file mode 100644
index 00000000..939e9e79
--- /dev/null
+++ b/test/benchmark/sysbench/thread_lat/bench_result.yaml
@@ -0,0 +1,11 @@
+alert:
+  bigger_is_better: false
+  threshold: 130%
+chart:
+  description: sysbench threads
+  legend: Average Execution Time per Thread on {system}
+  title: '[Threads] Threads performance'
+  unit: ms
+result_extraction:
+  result_index: NF
+  search_pattern: 'avg: +[0-9.]+'