diff --git a/.github/workflows/load-tests.yml b/.github/workflows/load-tests.yml
index 6c19491c97931..fabb24e2a148b 100644
--- a/.github/workflows/load-tests.yml
+++ b/.github/workflows/load-tests.yml
@@ -37,6 +37,7 @@ jobs:
           go-version: oldstable
       - run: make genoteltestbedcol
       - run: make oteltestbedcol
+      - run: GOOS=linux GOARCH=arm64 make oteltestbedcol
       - name: Upload Testbed Binaries
         uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
         with:
@@ -48,11 +49,11 @@ jobs:
       - run: ./.github/workflows/scripts/check-disk-space.sh
 
   loadtest:
-    runs-on: oracle-bare-metal-64cpu-1024gb-x86-64-ubuntu-24
     needs: [setup-environment]
     strategy:
       fail-fast: false
       matrix: ${{ fromJson(needs.setup-environment.outputs.loadtest_matrix) }}
+    runs-on: "${{ matrix.runner }}"
     steps:
       - uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
         with:
@@ -78,6 +79,7 @@ jobs:
       - name: Loadtest
         run: make -C testbed run-tests
         env:
+          RUN_TYPE: "${{ matrix.type == 'linux-arm64' && 'linux_arm64' || '' }}"
           TEST_ARGS: "-test.run=${{ matrix.test }}"
       - name: Set results filename
         if: ${{ failure() || success() }}
diff --git a/.github/workflows/scripts/setup_e2e_tests.sh b/.github/workflows/scripts/setup_e2e_tests.sh
index e56fa174377cf..9e74693cc8955 100755
--- a/.github/workflows/scripts/setup_e2e_tests.sh
+++ b/.github/workflows/scripts/setup_e2e_tests.sh
@@ -13,12 +13,12 @@ for i in "${!TEST_ARRAY[@]}"; do
     curr+="|${TEST_ARRAY[$i]}"
   else
     if [ -n "$curr" ] && (( i>1 )); then
-      MATRIX+=",{\"test\":\"$curr\"}"
+      MATRIX+=",{\"test\":\"$curr\",\"runner\":\"oracle-bare-metal-64cpu-1024gb-x86-64-ubuntu-24\",\"type\":\"linux-amd64\"},{\"test\":\"$curr\",\"runner\":\"cncf-ubuntu-32-128-arm\",\"type\":\"linux-arm64\"}"
     elif [ -n "$curr" ]; then
-      MATRIX+="{\"test\":\"$curr\"}"
+      MATRIX+="{\"test\":\"$curr\",\"runner\":\"oracle-bare-metal-64cpu-1024gb-x86-64-ubuntu-24\",\"type\":\"linux-amd64\"},{\"test\":\"$curr\",\"runner\":\"cncf-ubuntu-32-128-arm\",\"type\":\"linux-arm64\"}"
     fi
     curr="${TEST_ARRAY[$i]}"
   fi
 done
-MATRIX+=",{\"test\":\"$curr\"}]}"
+MATRIX+=",{\"test\":\"$curr\",\"runner\":\"oracle-bare-metal-64cpu-1024gb-x86-64-ubuntu-24\",\"type\":\"linux-amd64\"},{\"test\":\"$curr\",\"runner\":\"cncf-ubuntu-32-128-arm\",\"type\":\"linux-arm64\"}]}"
 echo "loadtest_matrix=$MATRIX" >> "$GITHUB_OUTPUT"
diff --git a/testbed/testbed/results.go b/testbed/testbed/results.go
index 361113da46d3e..5f2a684de526c 100644
--- a/testbed/testbed/results.go
+++ b/testbed/testbed/results.go
@@ -127,50 +127,53 @@ func (r *PerformanceResults) Add(_ string, result any) {
 	memoryChartName := fmt.Sprintf("%s - RAM (MiB)", testResult.testName)
 	droppedSpansChartName := fmt.Sprintf("%s - Dropped Span Count", testResult.testName)
 
-	r.benchmarkResults = append(r.benchmarkResults,
-		&benchmarkResult{
-			Name:  "cpu_percentage_avg",
-			Value: testResult.cpuPercentageAvg,
-			Unit:  "%",
-			Extra: cpuChartName,
-		},
-		&benchmarkResult{
-			Name:  "cpu_percentage_max",
-			Value: testResult.cpuPercentageMax,
-			Unit:  "%",
-			Extra: cpuChartName,
-		})
+	runType := ""
+	if t := os.Getenv("RUN_TYPE"); t != "" {
+		runType = fmt.Sprintf("_%s", t)
+	}
+
+	r.benchmarkResults = append(r.benchmarkResults, &benchmarkResult{
+		Name:  fmt.Sprintf("cpu_percentage_avg%s", runType),
+		Value: testResult.cpuPercentageAvg,
+		Unit:  "%",
+		Extra: cpuChartName,
+	})
+	r.benchmarkResults = append(r.benchmarkResults, &benchmarkResult{
+		Name:  fmt.Sprintf("cpu_percentage_max%s", runType),
+		Value: testResult.cpuPercentageMax,
+		Unit:  "%",
+		Extra: cpuChartName,
+	})
 	if testResult.cpuPercentageLimit > 0 {
 		r.benchmarkResults = append(r.benchmarkResults, &benchmarkResult{
-			Name:  "cpu_percentage_limit",
+			Name:  fmt.Sprintf("cpu_percentage_limit%s", runType),
 			Value: testResult.cpuPercentageLimit,
 			Unit:  "%",
 			Extra: cpuChartName,
 		})
 	}
-	r.benchmarkResults = append(r.benchmarkResults,
-		&benchmarkResult{
-			Name:  "ram_mib_avg",
-			Value: float64(testResult.ramMibAvg),
-			Unit:  "MiB",
-			Extra: memoryChartName,
-		},
-		&benchmarkResult{
-			Name:  "ram_mib_max",
-			Value: float64(testResult.ramMibMax),
-			Unit:  "MiB",
-			Extra: memoryChartName,
-		})
+	r.benchmarkResults = append(r.benchmarkResults, &benchmarkResult{
+		Name:  fmt.Sprintf("ram_mib_avg%s", runType),
+		Value: float64(testResult.ramMibAvg),
+		Unit:  "MiB",
+		Extra: memoryChartName,
+	})
+	r.benchmarkResults = append(r.benchmarkResults, &benchmarkResult{
+		Name:  fmt.Sprintf("ram_mib_max%s", runType),
+		Value: float64(testResult.ramMibMax),
+		Unit:  "MiB",
+		Extra: memoryChartName,
+	})
 	if testResult.ramMibLimit > 0 {
 		r.benchmarkResults = append(r.benchmarkResults, &benchmarkResult{
-			Name:  "ram_mib_limit",
+			Name:  fmt.Sprintf("ram_mib_limit%s", runType),
 			Value: float64(testResult.ramMibLimit),
 			Unit:  "MiB",
 			Extra: memoryChartName,
 		})
 	}
 	r.benchmarkResults = append(r.benchmarkResults, &benchmarkResult{
-		Name:  "dropped_span_count",
+		Name:  fmt.Sprintf("dropped_span_count%s", runType),
 		Value: float64(testResult.sentSpanCount - testResult.receivedSpanCount),
 		Unit:  "spans",
 		Extra: droppedSpansChartName,