Merge pull request GoogleCloudPlatform#606 from LujieDuan/lujieduan-u…

…pdate-dashboards-for-gpu-intergrations Update dashboards for gpu integrations
bombastictranz · Aug 15, 2023 · 46cc719 · 46cc719
2 parents 6e5c330 + ef93169
commit 46cc719
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 10 deletions.
diff --git a/dashboards/nvidia-gpu/metadata.yaml b/dashboards/nvidia-gpu/metadata.yaml
@@ -5,11 +5,6 @@ sample_dashboards:
     display_name: NVIDIA GPU Monitoring Overview (GCE & GKE)
     description: |-
       Displays GPU metrics for both GKE Nodes and GCE VMs.  GPU metrics for the GCE VMs require the Ops Agent to be installed.
-    related_integrations:
-      - id: nvml
-        platform: GCE
-      - id: dcgm
-        platform: GCE
   -
     category: NVIDIA GPUs
     id: nvidia-dcgm

diff --git a/dashboards/nvidia-gpu/nvidia-dcgm.json b/dashboards/nvidia-gpu/nvidia-dcgm.json
@@ -18,7 +18,7 @@
                 "plotType": "LINE",
                 "targetAxis": "Y1",
                 "timeSeriesQuery": {
-                  "timeSeriesQueryLanguage": "fetch gce_instance\n| metric 'workload.googleapis.com/dcgm.gpu.pipe_utilization'\n"
+                  "timeSeriesQueryLanguage": "fetch gce_instance\n| metric 'workload.googleapis.com/dcgm.gpu.profiling.pipe_utilization'\n"
                 }
               }
             ],
@@ -46,7 +46,7 @@
                 "plotType": "LINE",
                 "targetAxis": "Y1",
                 "timeSeriesQuery": {
-                  "timeSeriesQueryLanguage": "fetch gce_instance\n| metric 'workload.googleapis.com/dcgm.gpu.pcie_traffic_rate'\n| cast_units(\"By/s\")"
+                  "timeSeriesQueryLanguage": "fetch gce_instance\n| metric 'workload.googleapis.com/dcgm.gpu.profiling.pcie_traffic_rate'\n| cast_units(\"By/s\")"
                 }
               }
             ],
@@ -74,7 +74,7 @@
                 "plotType": "LINE",
                 "targetAxis": "Y1",
                 "timeSeriesQuery": {
-                  "timeSeriesQueryLanguage": "fetch gce_instance\n| metric 'workload.googleapis.com/dcgm.gpu.sm_utilization'\n"
+                  "timeSeriesQueryLanguage": "fetch gce_instance\n| metric 'workload.googleapis.com/dcgm.gpu.profiling.sm_utilization'\n"
                 }
               }
             ],
@@ -122,7 +122,7 @@
                 "plotType": "LINE",
                 "targetAxis": "Y1",
                 "timeSeriesQuery": {
-                  "timeSeriesQueryLanguage": "fetch gce_instance\n| metric 'workload.googleapis.com/dcgm.gpu.sm_occupancy'\n"
+                  "timeSeriesQueryLanguage": "fetch gce_instance\n| metric 'workload.googleapis.com/dcgm.gpu.profiling.sm_occupancy'\n"
                 }
               }
             ],
@@ -150,7 +150,7 @@
                 "plotType": "LINE",
                 "targetAxis": "Y1",
                 "timeSeriesQuery": {
-                  "timeSeriesQueryLanguage": "fetch gce_instance\n| metric 'workload.googleapis.com/dcgm.gpu.nvlink_traffic_rate'\n| cast_units(\"By/s\")"
+                  "timeSeriesQueryLanguage": "fetch gce_instance\n| metric 'workload.googleapis.com/dcgm.gpu.profiling.nvlink_traffic_rate'\n| cast_units(\"By/s\")"
                 }
               }
             ],