|
92 | 92 | "colorMode": "value",
|
93 | 93 | "graphMode": "none",
|
94 | 94 | "justifyMode": "auto",
|
95 |
| - "orientation": "auto", |
| 95 | + "orientation": "horizontal", |
96 | 96 | "reduceOptions": {
|
97 | 97 | "calcs": [
|
98 | 98 | "lastNotNull"
|
|
112 | 112 | },
|
113 | 113 | "disableTextWrap": false,
|
114 | 114 | "editorMode": "code",
|
115 |
| - "expr": "sum by(release)(envoy_cluster_membership_total{envoy_cluster_name=\"triton_grpc_service\", release=\"${release_name}\"})", |
| 115 | + "expr": "sum by(release)(envoy_cluster_membership_total{envoy_cluster_name=\"triton_grpc_service\", release=~\"${release_name}\"})", |
116 | 116 | "fullMetaSearch": false,
|
117 | 117 | "includeNullMetadata": true,
|
118 | 118 | "instant": false,
|
|
197 | 197 | "calcs": [],
|
198 | 198 | "displayMode": "list",
|
199 | 199 | "placement": "bottom",
|
200 |
| - "showLegend": false |
| 200 | + "showLegend": true |
201 | 201 | },
|
202 | 202 | "tooltip": {
|
203 | 203 | "mode": "multi",
|
|
211 | 211 | "uid": "prometheus"
|
212 | 212 | },
|
213 | 213 | "expr": "${server_load_metric}",
|
214 |
| - "legendFormat": "__auto", |
| 214 | + "legendFormat": "{{release}}", |
215 | 215 | "refId": "A"
|
216 | 216 | }
|
217 | 217 | ],
|
|
328 | 328 | "uid": "prometheus"
|
329 | 329 | },
|
330 | 330 | "editorMode": "code",
|
331 |
| - "expr": "sum by (release) ( rate(nv_inference_compute_infer_duration_us{release=\"${release_name}\"}[15s])) /sum by (release) ( (rate(nv_inference_exec_count{release=\"${release_name}\"}[15s]) * 1000) + 0.001)", |
| 331 | + "expr": "sum by (app) ( rate(nv_inference_compute_infer_duration_us{release=~\"${release_name}\"}[15s])) /sum by (app) ( (rate(nv_inference_exec_count{release=~\"${release_name}\"}[15s]) * 1000) + 0.001)", |
332 | 332 | "instant": false,
|
333 | 333 | "legendFormat": "Inference",
|
334 | 334 | "range": true,
|
|
340 | 340 | "uid": "prometheus"
|
341 | 341 | },
|
342 | 342 | "editorMode": "code",
|
343 |
| - "expr": "sum by (release) ( rate(nv_inference_queue_duration_us{release=\"${release_name}\"}[15s])) /sum by (release) ( (rate(nv_inference_exec_count{release=\"${release_name}\"}[15s]) * 1000) + 0.001)", |
| 343 | + "expr": "sum by (app) ( rate(nv_inference_queue_duration_us{release=~\"${release_name}\"}[15s])) /sum by (app) ( (rate(nv_inference_exec_count{release=~\"${release_name}\"}[15s]) * 1000) + 0.001)", |
344 | 344 | "hide": false,
|
345 | 345 | "instant": false,
|
346 | 346 | "legendFormat": "Queue",
|
|
353 | 353 | "uid": "prometheus"
|
354 | 354 | },
|
355 | 355 | "editorMode": "code",
|
356 |
| - "expr": "sum by (release) ( rate(nv_inference_compute_input_duration_us{release=\"${release_name}\"}[15s])) /sum by (release) ( (rate(nv_inference_exec_count{release=\"${release_name}\"}[15s]) * 1000) + 0.001)", |
| 356 | + "expr": "sum by (app) ( rate(nv_inference_compute_input_duration_us{release=~\"${release_name}\"}[15s])) /sum by (app) ( (rate(nv_inference_exec_count{release=~\"${release_name}\"}[15s]) * 1000) + 0.001)", |
357 | 357 | "hide": false,
|
358 | 358 | "instant": false,
|
359 | 359 | "legendFormat": "Input",
|
|
366 | 366 | "uid": "prometheus"
|
367 | 367 | },
|
368 | 368 | "editorMode": "code",
|
369 |
| - "expr": "sum by (release) ( rate(nv_inference_compute_output_duration_us{release=\"${release_name}\"}[15s])) /sum by (release) ( (rate(nv_inference_exec_count{release=\"${release_name}\"}[15s]) * 1000) + 0.001)", |
| 369 | + "expr": "sum by (app) ( rate(nv_inference_compute_output_duration_us{release=~\"${release_name}\"}[15s])) /sum by (app) ( (rate(nv_inference_exec_count{release=~\"${release_name}\"}[15s]) * 1000) + 0.001)", |
370 | 370 | "hide": false,
|
371 | 371 | "instant": false,
|
372 | 372 | "legendFormat": "Output",
|
|
379 | 379 | "uid": "prometheus"
|
380 | 380 | },
|
381 | 381 | "editorMode": "code",
|
382 |
| - "expr": " sum(\n rate(envoy_http_downstream_rq_time_sum{envoy_http_conn_manager_prefix=\"ingress_grpc\", release=\"${release_name}\"}[15s])\n /\n rate(envoy_http_downstream_rq_time_count{envoy_http_conn_manager_prefix=\"ingress_grpc\", release=\"${release_name}\"}[15s])\n ) by (release)", |
| 382 | + "expr": " sum(\n rate(envoy_http_downstream_rq_time_sum{envoy_http_conn_manager_prefix=\"ingress_grpc\", release=~\"${release_name}\"}[15s])\n /\n rate(envoy_http_downstream_rq_time_count{envoy_http_conn_manager_prefix=\"ingress_grpc\", release=~\"${release_name}\"}[15s])\n ) by (app)", |
383 | 383 | "hide": false,
|
384 | 384 | "instant": false,
|
385 | 385 | "legendFormat": "Total (measured at proxy)",
|
|
462 | 462 | "calcs": [],
|
463 | 463 | "displayMode": "list",
|
464 | 464 | "placement": "bottom",
|
465 |
| - "showLegend": false |
| 465 | + "showLegend": true |
466 | 466 | },
|
467 | 467 | "tooltip": {
|
468 | 468 | "mode": "multi",
|
|
477 | 477 | },
|
478 | 478 | "disableTextWrap": false,
|
479 | 479 | "editorMode": "code",
|
480 |
| - "expr": "sum by(release)(envoy_cluster_membership_total{envoy_cluster_name=\"triton_grpc_service\", release=\"${release_name}\"})", |
| 480 | + "expr": "sum by(release)(envoy_cluster_membership_total{envoy_cluster_name=\"triton_grpc_service\", release=~\"${release_name}\"})", |
481 | 481 | "fullMetaSearch": false,
|
482 | 482 | "includeNullMetadata": true,
|
483 | 483 | "instant": false,
|
|
909 | 909 | "calcs": [],
|
910 | 910 | "displayMode": "list",
|
911 | 911 | "placement": "right",
|
912 |
| - "showLegend": false |
| 912 | + "showLegend": true |
913 | 913 | },
|
914 | 914 | "tooltip": {
|
915 | 915 | "mode": "multi",
|
|
925 | 925 | },
|
926 | 926 | "editorMode": "code",
|
927 | 927 | "exemplar": false,
|
928 |
| - "expr": "sum (rate(nv_inference_count{release=\"${release_name}\"}[15s]))by(release)", |
| 928 | + "expr": "sum (rate(nv_inference_count{release=~\"${release_name}\"}[15s]))by(release)", |
929 | 929 | "instant": false,
|
930 | 930 | "interval": "",
|
931 | 931 | "legendFormat": "{{ release }}",
|
|
1022 | 1022 | },
|
1023 | 1023 | "editorMode": "code",
|
1024 | 1024 | "exemplar": false,
|
1025 |
| - "expr": "sum (rate(nv_inference_count{release=\"${release_name}\"}[15s]))by(pod)", |
| 1025 | + "expr": "sum (rate(nv_inference_count{release=~\"${release_name}\"}[15s]))by(pod)", |
1026 | 1026 | "instant": false,
|
1027 | 1027 | "interval": "",
|
1028 | 1028 | "legendFormat": "{{ pod }}",
|
|
1119 | 1119 | },
|
1120 | 1120 | "editorMode": "code",
|
1121 | 1121 | "exemplar": false,
|
1122 |
| - "expr": "sum (rate(nv_inference_count{release=\"${release_name}\"}[15s]))by(model)", |
| 1122 | + "expr": "sum (rate(nv_inference_count{release=~\"${release_name}\"}[15s]))by(model)", |
1123 | 1123 | "instant": false,
|
1124 | 1124 | "interval": "",
|
1125 | 1125 | "legendFormat": "{{ model }}",
|
|
0 commit comments