-
Notifications
You must be signed in to change notification settings - Fork 48
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Merge pull request #224 from israel-hdez/release-0.12.0-main-sync
Release 0.12.0 main sync
- Loading branch information
Showing
23 changed files
with
715 additions
and
221 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,76 @@ | ||
| apiVersion: template.openshift.io/v1 | ||
| kind: Template | ||
| metadata: | ||
| labels: | ||
| opendatahub.io/dashboard: 'true' | ||
| opendatahub.io/ootb: 'true' | ||
| annotations: | ||
| description: Caikit is an AI toolkit that enables users to manage models through a set of developer friendly APIs. It provides a consistent format for creating and using AI models against a wide variety of data domains and tasks. | ||
| openshift.io/provider-display-name: Red Hat, Inc. | ||
| tags: rhods,rhoai,kserve,servingruntime | ||
| template.openshift.io/documentation-url: https://github.com/opendatahub-io/caikit-nlp | ||
| template.openshift.io/long-description: This template defines resources needed to deploy caikit-standalone-serving servingruntime with Red Hat Data Science KServe for LLM model | ||
| template.openshift.io/support-url: https://access.redhat.com | ||
| opendatahub.io/modelServingSupport: '["single"]' | ||
| opendatahub.io/apiProtocol: 'REST' | ||
| name: caikit-standalone-serving-template | ||
| objects: | ||
| - apiVersion: serving.kserve.io/v1alpha1 | ||
| kind: ServingRuntime | ||
| metadata: | ||
| name: caikit-standalone-runtime | ||
| annotations: | ||
| openshift.io/display-name: Caikit Standalone ServingRuntime for KServe | ||
| opendatahub.io/recommended-accelerators: '["nvidia.com/gpu"]' | ||
| labels: | ||
| opendatahub.io/dashboard: 'true' | ||
| spec: | ||
| annotations: | ||
| prometheus.io/port: '8086' | ||
| prometheus.io/path: /metrics | ||
| multiModel: false | ||
| supportedModelFormats: | ||
| - autoSelect: true | ||
| name: caikit | ||
| containers: | ||
| - name: kserve-container | ||
| image: $(caikit-standalone-image) | ||
| command: | ||
| - python | ||
| - '-m' | ||
| - caikit.runtime | ||
| env: | ||
| - name: RUNTIME_LOCAL_MODELS_DIR | ||
| value: /mnt/models | ||
| - name: HF_HOME | ||
| value: /tmp/hf_home | ||
| - name: RUNTIME_GRPC_ENABLED | ||
| value: 'false' | ||
| - name: RUNTIME_HTTP_ENABLED | ||
| value: 'true' | ||
| ports: | ||
| - containerPort: 8080 | ||
| protocol: TCP | ||
| readinessProbe: | ||
| exec: | ||
| command: | ||
| - python | ||
| - -m | ||
| - caikit_health_probe | ||
| - readiness | ||
| initialDelaySeconds: 5 | ||
| livenessProbe: | ||
| exec: | ||
| command: | ||
| - python | ||
| - -m | ||
| - caikit_health_probe | ||
| - liveness | ||
| initialDelaySeconds: 5 | ||
| startupProbe: | ||
| httpGet: | ||
| port: 8080 | ||
| path: /health | ||
| # Allow 12 mins to start | ||
| failureThreshold: 24 | ||
| periodSeconds: 30 |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -55,4 +55,4 @@ objects: | |
| value: 'true' | ||
| ports: | ||
| - containerPort: 8080 | ||
| protocol: TCP | ||
| protocol: TCP | ||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,55 @@ | ||
| { | ||
| "metrics": { | ||
| "supported": "true", | ||
| "config": [ | ||
| { | ||
| "title": "Number of requests", | ||
| "type": "REQUEST_COUNT", | ||
| "queries": [ | ||
| { | ||
| "title": "Number of successful incoming requests", | ||
| "query": "sum(increase(predict_rpc_count_total{namespace='${NAMESPACE}',code='OK',model_id='${MODEL_NAME}'}[${RATE_INTERVAL}]))" | ||
| }, | ||
| { | ||
| "title": "Number of failed incoming requests", | ||
| "query": "sum(increase(predict_rpc_count_total{namespace='${NAMESPACE}',code!='OK',model_id='${MODEL_NAME}'}[${RATE_INTERVAL}]))" | ||
| } | ||
| ] | ||
| }, | ||
| { | ||
| "title": "Average response time (ms)", | ||
| "type": "MEAN_LATENCY", | ||
| "queries": [ | ||
| { | ||
| "title": "Average inference latency", | ||
| "query": "sum by (model_id) (rate(predict_caikit_library_duration_seconds_sum{namespace='${NAMESPACE}',model_id='${MODEL_NAME}'}[1m])) / sum by (model_id) (rate(predict_caikit_library_duration_seconds_count{namespace='${NAMESPACE}',model_id='${MODEL_NAME}'}[${RATE_INTERVAL}]))" | ||
| }, | ||
| { | ||
| "title": "Average e2e latency", | ||
| "query": "sum by (model_id) (rate(caikit_core_load_model_duration_seconds_sum{namespace='${NAMESPACE}',model_id='${MODEL_NAME}'}[1m]) + rate(predict_caikit_library_duration_seconds_sum{namespace='${NAMESPACE}',model_id='${MODEL_NAME}'}[1m])) / sum by (model_id) (rate(caikit_core_load_model_duration_seconds_count{namespace='${NAMESPACE}',model_id='${MODEL_NAME}'}[${RATE_INTERVAL}]) + rate(predict_caikit_library_duration_seconds_count{namespace='${NAMESPACE}',model_id='${MODEL_NAME}'}[${RATE_INTERVAL}]))" | ||
| } | ||
| ] | ||
| }, | ||
| { | ||
| "title": "CPU utilization %", | ||
| "type": "CPU_USAGE", | ||
| "queries": [ | ||
| { | ||
| "title": "CPU usage", | ||
| "query": "sum(node_namespace_pod_container:container_cpu_usage_seconds_total:sum_irate{namespace='${NAMESPACE}'}* on(namespace,pod) group_left(workload, workload_type) namespace_workload_pod:kube_pod_owner:relabel{namespace='${NAMESPACE}', workload=~'${MODEL_NAME}-predictor-.*', workload_type=~'deployment'}) by (pod)" | ||
| } | ||
| ] | ||
| }, | ||
| { | ||
| "title": "Memory utilization %", | ||
| "type": "MEMORY_USAGE", | ||
| "queries": [ | ||
| { | ||
| "title": "Memory usage", | ||
| "query": "sum(container_memory_working_set_bytes{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}) by (pod)" | ||
| } | ||
| ] | ||
| } | ||
| ] | ||
| } | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,55 @@ | ||
| { | ||
| "metrics": { | ||
| "supported": "true", | ||
| "config": [ | ||
| { | ||
| "title": "Number of requests", | ||
| "type": "REQUEST_COUNT", | ||
| "queries": [ | ||
| { | ||
| "title": "Number of successful incoming requests", | ||
| "query": "sum(increase(ovms_requests_success{namespace='${NAMESPACE}',name='${MODEL_NAME}'}[${RATE_INTERVAL}]))" | ||
| }, | ||
| { | ||
| "title": "Number of failed incoming requests", | ||
| "query": "sum(increase(ovms_requests_fail{namespace='${NAMESPACE}',name='${MODEL_NAME}'}[${RATE_INTERVAL}]))" | ||
| } | ||
| ] | ||
| }, | ||
| { | ||
| "title": "Average response time (ms)", | ||
| "type": "MEAN_LATENCY", | ||
| "queries": [ | ||
| { | ||
| "title": "Average inference latency", | ||
| "query": "sum by (name) (rate(ovms_inference_time_us_sum{namespace='${NAMESPACE}', name='${MODEL_NAME}'}[1m])) / sum by (name) (rate(ovms_inference_time_us_count{namespace='${NAMESPACE}', name='${MODEL_NAME}'}[{RATE_INTERVAL}]))" | ||
| }, | ||
| { | ||
| "title": "Average e2e latency", | ||
| "query": "sum by (name) (rate(ovms_request_time_us_sum{name='${MODEL_NAME}'}[1m])) / sum by (name) (rate(ovms_request_time_us_count{name='${MODEL_NAME}'}[{RATE_INTERVAL}]))" | ||
| } | ||
| ] | ||
| }, | ||
| { | ||
| "title": "CPU utilization %", | ||
| "type": "CPU_USAGE", | ||
| "queries": [ | ||
| { | ||
| "title": "CPU usage", | ||
| "query": "sum(node_namespace_pod_container:container_cpu_usage_seconds_total:sum_irate{namespace='${NAMESPACE}'}* on(namespace,pod) group_left(workload, workload_type) namespace_workload_pod:kube_pod_owner:relabel{namespace='${NAMESPACE}', workload=~'${MODEL_NAME}-predictor-.*', workload_type=~'deployment'}) by (pod)" | ||
| } | ||
| ] | ||
| }, | ||
| { | ||
| "title": "Memory utilization %", | ||
| "type": "MEMORY_USAGE", | ||
| "queries": [ | ||
| { | ||
| "title": "Memory usage", | ||
| "query": "sum(container_memory_working_set_bytes{namespace='$(MODEL_NAMESPACE)', pod=~'${MODEL_NAME}-predictor-.*'}) by (pod)" | ||
| } | ||
| ] | ||
| } | ||
| ] | ||
| } | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,55 @@ | ||
| { | ||
| "metrics": { | ||
| "supported": "true", | ||
| "config": [ | ||
| { | ||
| "title": "Number of requests", | ||
| "type": "REQUEST_COUNT", | ||
| "queries": [ | ||
| { | ||
| "title": "Number of successful incoming requests", | ||
| "query": "sum(increase(tgi_request_success{namespace=${NAMESPACE}, pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}]))" | ||
| }, | ||
| { | ||
| "title": "Number of failed incoming requests", | ||
| "query": "sum(increase(tgi_request_failure{namespace=${NAMESPACE}, pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}]))" | ||
| } | ||
| ] | ||
| }, | ||
| { | ||
| "title": "Average response time (ms)", | ||
| "type": "MEAN_LATENCY", | ||
| "queries": [ | ||
| { | ||
| "title": "Average inference latency", | ||
| "query": "sum by (pod) (rate(tgi_request_inference_duration_sum{namespace=${NAMESPACE}, pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}])) / sum by (pod) (rate(tgi_request_inference_duration_count{namespace=${NAMESPACE}, pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}])) " | ||
| }, | ||
| { | ||
| "title": "Average e2e latency", | ||
| "query": "sum by (pod) (rate(tgi_request_duration_sum{namespace=${NAMESPACE}, pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}])) / sum by (pod) (rate(tgi_request_duration_count{namespace=${NAMESPACE}, pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}]))" | ||
| } | ||
| ] | ||
| }, | ||
| { | ||
| "title": "CPU utilization %", | ||
| "type": "CPU_USAGE", | ||
| "queries": [ | ||
| { | ||
| "title": "CPU usage", | ||
| "query": "sum(node_namespace_pod_container:container_cpu_usage_seconds_total:sum_irate{namespace='${NAMESPACE}'}* on(namespace,pod) group_left(workload, workload_type) namespace_workload_pod:kube_pod_owner:relabel{namespace='${NAMESPACE}', workload=~'${MODEL_NAME}-predictor-.*', workload_type=~'deployment'}) by (pod)" | ||
| } | ||
| ] | ||
| }, | ||
| { | ||
| "title": "Memory utilization %", | ||
| "type": "MEMORY_USAGE", | ||
| "queries": [ | ||
| { | ||
| "title": "Memory usage", | ||
| "query": "sum(container_memory_working_set_bytes{namespace='$(MODEL_NAMESPACE)', pod=~'${MODEL_NAME}-predictor-.*'}) by (pod)" | ||
| } | ||
| ] | ||
| } | ||
| ] | ||
| } | ||
| } |
Oops, something went wrong.