diff --git a/dashboards/gpu/gpu-efficiency.json b/dashboards/gpu/gpu-efficiency.json index e2e8c105..3bd59bc0 100644 --- a/dashboards/gpu/gpu-efficiency.json +++ b/dashboards/gpu/gpu-efficiency.json @@ -166,12 +166,12 @@ "id": 4, "targets": [ { - "expr": "avg(gpu:power_throttle_fraction:rate5m{namespace=~\"$namespace\"})", + "expr": "avg(gpu:power_throttle_fraction:rate5m)", "refId": "A" } ], "title": "Avg Power Throttling", - "description": "Fraction of time GPUs hit the TDP cap and lose performance. \u003e5% means tenants underutilize billed FLOPS.", + "description": "Fraction of time GPUs hit the TDP cap and lose performance. \u003e5% means tenants underutilize billed FLOPS. Cluster-wide — rule aggregates by (Hostname, gpu, UUID) and drops namespace label.", "transparent": false, "datasource": { "type": "prometheus", @@ -658,7 +658,7 @@ } ], "title": "Power throttle fraction per GPU", - "description": "Fraction of time the GPU was power-throttled. 1.0 = always throttled.", + "description": "Fraction of time the GPU was power-throttled. 1.0 = always throttled. Cluster-wide — rule aggregates by (Hostname, gpu, UUID) and drops namespace label.", "transparent": false, "datasource": { "type": "prometheus", @@ -729,7 +729,7 @@ } ], "title": "Thermal throttle fraction per GPU", - "description": "Fraction of time the GPU was thermal-throttled.", + "description": "Fraction of time the GPU was thermal-throttled. Cluster-wide — rule aggregates by (Hostname, gpu, UUID) and drops namespace label.", "transparent": false, "datasource": { "type": "prometheus",