bstein/titan-iac
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

monitoring: source gpu pie from limits and node nets

Browse Source
This commit is contained in:
Brad Stein 2025-11-18 01:01:10 -03:00
parent 5144bbe1f2
commit ec76563a86
3 changed files with 10 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
scripts/render_dashboards.py
View File

@ -178,7 +178,7 @@ def namespace_ram_share_expr():
def namespace_gpu_share_expr():
selected = f"(( {NAMESPACE_GPU_RAW} ) or on(namespace) ( {NAMESPACE_COMBINED_FILTER} * 0)) and on(namespace) ( {NAMESPACE_COMBINED_FILTER} )"
selected = f"( {NAMESPACE_GPU_RAW} ) and on(namespace) ( {NAMESPACE_COMBINED_FILTER} )"
total = f"clamp_min(sum( {NAMESPACE_GPU_RAW} ), 1)"
return f"100 * ( {selected} ) / {total}"
@ -223,8 +223,9 @@ NAMESPACE_RAM_RAW = (
'sum(container_memory_working_set_bytes{namespace!="",pod!="",container!=""}) by (namespace)'
)
NAMESPACE_GPU_RAW = (
'sum(kube_pod_container_resource_requests{namespace!="",resource="nvidia.com/gpu"} '
'or kube_pod_resource_request{namespace!="",resource="nvidia.com/gpu"}) by (namespace)'
'sum((kube_pod_container_resource_requests{namespace!="",resource="nvidia.com/gpu"}'
' or kube_pod_container_resource_limits{namespace!="",resource="nvidia.com/gpu"})'
') by (namespace)'
)
NAMESPACE_GPU_WEIGHT = (
"(( "

6
services/monitoring/dashboards/atlas-overview.json
View File

@ -722,7 +722,7 @@
},
"targets": [
{
"expr": "100 * ( ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ), 1)",
"expr": "100 * ( ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum((kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_container_resource_limits{namespace!=\"\",resource=\"nvidia.com/gpu\"})) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}
@ -764,7 +764,7 @@
},
"targets": [
{
"expr": "100 * ( (( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) * 0)) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ), 1)",
"expr": "100 * ( ( sum((kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_container_resource_limits{namespace!=\"\",resource=\"nvidia.com/gpu\"})) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum((kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_container_resource_limits{namespace!=\"\",resource=\"nvidia.com/gpu\"})) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum((kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_container_resource_limits{namespace!=\"\",resource=\"nvidia.com/gpu\"})) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}
@ -806,7 +806,7 @@
},
"targets": [
{
"expr": "100 * ( ( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ), 1)",
"expr": "100 * ( ( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum((kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_container_resource_limits{namespace!=\"\",resource=\"nvidia.com/gpu\"})) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}

6
services/monitoring/grafana-dashboard-overview.yaml
View File

@ -731,7 +731,7 @@ data:
},
"targets": [
{
"expr": "100 * ( ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ), 1)",
"expr": "100 * ( ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum((kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_container_resource_limits{namespace!=\"\",resource=\"nvidia.com/gpu\"})) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}
@ -773,7 +773,7 @@ data:
},
"targets": [
{
"expr": "100 * ( (( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) * 0)) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ), 1)",
"expr": "100 * ( ( sum((kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_container_resource_limits{namespace!=\"\",resource=\"nvidia.com/gpu\"})) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum((kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_container_resource_limits{namespace!=\"\",resource=\"nvidia.com/gpu\"})) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum((kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_container_resource_limits{namespace!=\"\",resource=\"nvidia.com/gpu\"})) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}
@ -815,7 +815,7 @@ data:
},
"targets": [
{
"expr": "100 * ( ( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum(kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_resource_request{namespace!=\"\",resource=\"nvidia.com/gpu\"}) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ), 1)",
"expr": "100 * ( ( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ) and on(namespace) ( topk(10, (sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace)) + (sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) / 1e9) + ((( sum((kube_pod_container_resource_requests{namespace!=\"\",resource=\"nvidia.com/gpu\"} or kube_pod_container_resource_limits{namespace!=\"\",resource=\"nvidia.com/gpu\"})) by (namespace) ) or on(namespace) ( sum(rate(container_cpu_usage_seconds_total{namespace!=\"\",pod!=\"\",container!=\"\"}[5m])) by (namespace) * 0)) * 10)) ) ) / clamp_min(sum( sum(container_memory_working_set_bytes{namespace!=\"\",pod!=\"\",container!=\"\"}) by (namespace) ), 1)",
"refId": "A",
"legendFormat": "{{namespace}}"
}