Skip to content

Update integ tests for new Tensor Core Utilization and GPU Metrics #582

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from
Closed

Update integ tests for new Tensor Core Utilization and GPU Metrics #582

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
83 changes: 44 additions & 39 deletions test/gpu/nvidia_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,71 +18,76 @@ import (
const (
gpuMetricIndicator = "_gpu_"

containerMemTotal = "container_gpu_memory_total"
containerMemUsed = "container_gpu_memory_used"
containerPower = "container_gpu_power_draw"
containerTemp = "container_gpu_temperature"
containerUtil = "container_gpu_utilization"
containerMemUtil = "container_gpu_memory_utilization"
podMemTotal = "pod_gpu_memory_total"
podMemUsed = "pod_gpu_memory_used"
podPower = "pod_gpu_power_draw"
podTemp = "pod_gpu_temperature"
podUtil = "pod_gpu_utilization"
podMemUtil = "pod_gpu_memory_utilization"
podLimit = "pod_gpu_limit"
podRequest = "pod_gpu_request"
podCountTotal = "pod_gpu_usage_total"
podReserved = "pod_gpu_reserved_capacity"
nodeMemTotal = "node_gpu_memory_total"
nodeMemUsed = "node_gpu_memory_used"
nodePower = "node_gpu_power_draw"
nodeTemp = "node_gpu_temperature"
nodeUtil = "node_gpu_utilization"
nodeMemUtil = "node_gpu_memory_utilization"
nodeCountTotal = "node_gpu_usage_total"
nodeCountLimit = "node_gpu_limit"
nodeReserved = "node_gpu_reserved_capacity"
containerMemTotal = "container_gpu_memory_total"
containerMemUsed = "container_gpu_memory_used"
containerPower = "container_gpu_power_draw"
containerTemp = "container_gpu_temperature"
containerUtil = "container_gpu_utilization"
containerMemUtil = "container_gpu_memory_utilization"
containerTensorUtil = "container_gpu_tensor_core_utilization"
podMemTotal = "pod_gpu_memory_total"
podMemUsed = "pod_gpu_memory_used"
podPower = "pod_gpu_power_draw"
podTemp = "pod_gpu_temperature"
podUtil = "pod_gpu_utilization"
podMemUtil = "pod_gpu_memory_utilization"
podTensorUtil = "pod_gpu_tensor_core_utilization"
podLimit = "pod_gpu_limit"
podRequest = "pod_gpu_request"
podCountTotal = "pod_gpu_usage_total"
podReserved = "pod_gpu_reserved_capacity"
nodeMemTotal = "node_gpu_memory_total"
nodeMemUsed = "node_gpu_memory_used"
nodePower = "node_gpu_power_draw"
nodeTemp = "node_gpu_temperature"
nodeUtil = "node_gpu_utilization"
nodeMemUtil = "node_gpu_memory_utilization"
nodeTensorUtil = "node_gpu_tensor_core_utilization"
nodeCountTotal = "node_gpu_usage_total"
nodeCountLimit = "node_gpu_limit"
nodeReserved = "node_gpu_reserved_capacity"
nodeUnreserved = "node_gpu_unreserved_capacity"
nodeAvailable = "node_gpu_available_capacity"
)

var useE2EMetrics = flag.Bool("useE2EMetrics", false, "Use E2E metrics mapping which uses latest build CWA")

var expectedDimsToMetricsIntegTest = map[string][]string{
"ClusterName": {
containerMemTotal, containerMemUsed, containerPower, containerTemp, containerUtil, containerMemUtil,
podMemTotal, podMemUsed, podPower, podTemp, podUtil, podMemUtil,
nodeMemTotal, nodeMemUsed, nodePower, nodeTemp, nodeUtil, nodeMemUtil,
containerMemTotal, containerMemUsed, containerPower, containerTemp, containerUtil, containerMemUtil, containerTensorUtil,
podMemTotal, podMemUsed, podPower, podTemp, podUtil, podMemUtil, podTensorUtil,
nodeMemTotal, nodeMemUsed, nodePower, nodeTemp, nodeUtil, nodeMemUtil, nodeTensorUtil,
},
"ClusterName-Namespace": {
podMemTotal, podMemUsed, podPower, podTemp, podUtil, podMemUtil,
podMemTotal, podMemUsed, podPower, podTemp, podUtil, podMemUtil, podTensorUtil,
},
//"ClusterName-Namespace-Service": {
// podMemTotal, podMemUsed, podPower, podTemp, podUtil, podMemUtil,
//},
"ClusterName-Namespace-PodName": {
podMemTotal, podMemUsed, podPower, podTemp, podUtil, podMemUtil,
podMemTotal, podMemUsed, podPower, podTemp, podUtil, podMemUtil, podTensorUtil,
},
"ClusterName-ContainerName-Namespace-PodName": {
containerMemTotal, containerMemUsed, containerPower, containerTemp, containerUtil, containerMemUtil,
containerMemTotal, containerMemUsed, containerPower, containerTemp, containerUtil, containerMemUtil, containerTensorUtil,
},
"ClusterName-ContainerName-FullPodName-Namespace-PodName": {
containerMemTotal, containerMemUsed, containerPower, containerTemp, containerUtil, containerMemUtil,
containerMemTotal, containerMemUsed, containerPower, containerTemp, containerUtil, containerMemUtil, containerTensorUtil,
},
"ClusterName-ContainerName-FullPodName-GpuDevice-Namespace-PodName": {
containerMemTotal, containerMemUsed, containerPower, containerTemp, containerUtil, containerMemUtil,
containerMemTotal, containerMemUsed, containerPower, containerTemp, containerUtil, containerMemUtil, containerTensorUtil,
},
"ClusterName-FullPodName-Namespace-PodName": {
podMemTotal, podMemUsed, podPower, podTemp, podUtil, podMemUtil,
podMemTotal, podMemUsed, podPower, podTemp, podUtil, podMemUtil, podTensorUtil,
},
"ClusterName-FullPodName-GpuDevice-Namespace-PodName": {
podMemTotal, podMemUsed, podPower, podTemp, podUtil, podMemUtil,
podMemTotal, podMemUsed, podPower, podTemp, podUtil, podMemUtil, podTensorUtil,
},
"ClusterName-InstanceId-NodeName": {
nodeMemTotal, nodeMemUsed, nodePower, nodeTemp, nodeUtil, nodeMemUtil,
nodeMemTotal, nodeMemUsed, nodePower, nodeTemp, nodeUtil, nodeMemUtil, nodeTensorUtil,
//nodeCountTotal, nodeCountRequest, nodeCountLimit,
},
"ClusterName-GpuDevice-InstanceId-InstanceType-NodeName": {
nodeMemTotal, nodeMemUsed, nodePower, nodeTemp, nodeUtil, nodeMemUtil,
nodeMemTotal, nodeMemUsed, nodePower, nodeTemp, nodeUtil, nodeMemUtil, nodeTensorUtil,
},
}

Expand All @@ -99,10 +104,10 @@ func (t *NvidiaTestRunner) Validate() status.TestGroupResult {
expectedDimsToMetrics := expectedDimsToMetricsIntegTest
if *useE2EMetrics {
// add GPU count metrics
expectedDimsToMetricsIntegTest["ClusterName"] = append(expectedDimsToMetricsIntegTest["ClusterName"], podReserved, podRequest, podCountTotal, podLimit, nodeCountTotal, nodeCountLimit, nodeReserved)
expectedDimsToMetricsIntegTest["ClusterName"] = append(expectedDimsToMetricsIntegTest["ClusterName"], podReserved, podRequest, podCountTotal, podLimit, nodeCountTotal, nodeCountLimit, nodeReserved, nodeUnreserved, nodeAvailable)
expectedDimsToMetricsIntegTest["ClusterName-Namespace-PodName"] = append(expectedDimsToMetricsIntegTest["ClusterName-Namespace-PodName"], podCountTotal, podRequest, podReserved, podLimit)
expectedDimsToMetricsIntegTest["ClusterName-FullPodName-Namespace-PodName"] = append(expectedDimsToMetricsIntegTest["ClusterName-FullPodName-Namespace-PodName"], podCountTotal, podRequest, podReserved, podLimit)
expectedDimsToMetricsIntegTest["ClusterName-InstanceId-NodeName"] = append(expectedDimsToMetricsIntegTest["ClusterName-InstanceId-NodeName"], nodeCountLimit, nodeCountTotal, nodeReserved)
expectedDimsToMetricsIntegTest["ClusterName-InstanceId-NodeName"] = append(expectedDimsToMetricsIntegTest["ClusterName-InstanceId-NodeName"], nodeCountLimit, nodeCountTotal, nodeReserved, nodeUnreserved, nodeAvailable)
}
testResults = append(testResults, metric.ValidateMetrics(t.env, gpuMetricIndicator, expectedDimsToMetrics)...)
testResults = append(testResults, metric.ValidateLogs(t.env))
Expand Down
1 change: 1 addition & 0 deletions test/metric_value_benchmark/eks_resources/test_schemas/container_gpu.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,7 @@
"container_gpu_temperature": {},
"container_gpu_utilization": {},
"container_gpu_memory_utilization": {},
"container_gpu_tensor_core_utilization": {},
"Service":{}
},
"required": [
Expand Down
3 changes: 3 additions & 0 deletions test/metric_value_benchmark/eks_resources/test_schemas/node_gpu.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,9 +26,12 @@
"node_gpu_temperature": {},
"node_gpu_utilization": {},
"node_gpu_memory_utilization": {},
"node_gpu_tensor_core_utilization": {},
"node_gpu_total": {},
"node_gpu_request": {},
"node_gpu_list": {},
"node_gpu_unreserved_capacity": {},
"node_gpu_available_capacity": {},
"Service":{}
},
"required": [
Expand Down
1 change: 1 addition & 0 deletions test/metric_value_benchmark/eks_resources/test_schemas/pod_gpu.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,7 @@
"pod_gpu_temperature": {},
"pod_gpu_utilization": {},
"pod_gpu_memory_utilization": {},
"pod_gpu_tensor_core_utilization": {},
"pod_gpu_total": {},
"pod_gpu_request": {},
"pod_gpu_list": {},
Expand Down
Loading