8000 Remove default resources requests and limits - they should be set exp… by evilr00t · Pull Request #4448 · kserve/kserve · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Remove default resources requests and limits - they should be set exp… #4448

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 3 additions & 9 deletions charts/kserve-resources/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -46,27 +46,21 @@ $ helm install kserve oci://ghcr.io/kserve/charts/kserve --version v0.15.2
| kserve.controller.nodeSelector | object | `{}` | The nodeSelector on Pods tells Kubernetes to schedule Pods on the nodes with matching labels. For more information, see [Assigning Pods to Nodes](https://kubernetes.io/docs/concepts/scheduling-eviction/assign-pod-node/). |
| kserve.controller.podAnnotations | object | `{}` | Optional additional annotations to add to the controller Pods. |
| kserve.controller.podLabels | object | `{}` | Optional additional labels to add to the controller Pods. |
| kserve.controller.rbacProxy.resources.limits.cpu | string | `"100m"` | |
| kserve.controller.rbacProxy.resources.limits.memory | string | `"300Mi"` | |
| kserve.controller.rbacProxy.resources.requests.cpu | string | `"100m"` | |
| kserve.controller.rbacProxy.resources.requests.memory | string | `"300Mi"` | |
| kserve.controller.rbacProxy.resources | object | `{}` | |
| kserve.controller.rbacProxy.securityContext.allowPrivilegeEscalation | bool | `false` | |
| kserve.controller.rbacProxy.securityContext.capabilities.drop[0] | string | `"ALL"` | |
| kserve.controller.rbacProxy.securityContext.privileged | bool | `false` | |
| kserve.controller.rbacProxy.securityContext.readOnlyRootFilesystem | bool | `true` | |
| kserve.controller.rbacProxy.securityContext.runAsNonRoot | bool | `true` | |
| kserve.controller.rbacProxyImage | string | `"quay.io/brancz/kube-rbac-proxy:v0.18.0"` | KServe controller manager rbac proxy contrainer image |
| kserve.controller.resources | object | `{"limits":{"cpu":"100m","memory":"300Mi"},"requests":{"cpu":"100m","memory":"300Mi"}}` | Resources to provide to the kserve controller pod. For example: requests: cpu: 10m memory: 32Mi For more information, see [Resource Management for Pods and Containers](https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/). |
| kserve.controller.resources | object | `{}` | Resources to provide to the kserve controller pod. For example: requests: cpu: 10m memory: 32Mi For more information, see [Resource Management for Pods and Containers](https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/). |
| kserve.controller.securityContext | object | `{"runAsNonRoot":true}` | Pod Security Context. For more information, see [Configure a Security Context for a Pod or Container](https://kubernetes.io/docs/tasks/configure-pod-container/security-context/). |
| kserve.controller.serviceAnnotations | object | `{}` | Optional additional annotations to add to the controller service. |
| kserve.controller.tag | string | `"v0.15.2"` | KServe controller contrainer image tag. |
| kserve.controller.tolerations | list | `[]` | A list of Kubernetes Tolerations, if required. For more information, see [Toleration v1 core](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.27/#toleration-v1-core). For example: tolerations: - key: foo.bar.com/role operator: Equal value: master effect: NoSchedule |
| kserve.controller.topologySpreadConstraints | list | `[]` | A list of Kubernetes TopologySpreadConstraints, if required. For more information, see [Topology spread constraint v1 core](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.27/#topologyspreadconstraint-v1-core For example: topologySpreadConstraints: - maxSkew: 2 topologyKey: topology.kubernetes.io/zone whenUnsatisfiable: ScheduleAnyway labelSelector: matchLabels: app.kubernetes.io/instance: kserve-controller-manager app.kubernetes.io/component: controller |
| kserve.controller.webhookServiceAnnotations | object | `{}` | Optional additional annotations to add to the webhook service. |
| kserve.inferenceservice.resources.limits.cpu | string | `"1"` | |
| kserve.inferenceservice.resources.limits.memory | string | `"2Gi"` | |
| kserve.inferenceservice.resources.requests.cpu | string | `"1"` | |
| kserve.inferenceservice.resources.requests.memory | string | `"2Gi"` | |
| kserve.inferenceservice.resources | object | `{}` | The default InferenceService resources limit. |
| kserve.localmodel.agent.affinity | object | `{}` | |
| kserve.localmodel.agent.hostPath | string | `"/mnt/models"` | |
| kserve.localmodel.agent.image | string | `"kserve/kserve-localmodelnode-agent"` | |
Expand Down
25 changes: 19 additions & 6 deletions charts/kserve-resources/templates/configmap.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -683,13 +683,26 @@ data:
}

inferenceService: |-
{{- $resourceConfig := dict }}
{{- $kserve := .Values.kserve | default dict }}
{{- $inferenceservice := $kserve.inferenceservice | default dict }}
{{- $resources := $inferenceservice.resources | default dict }}
{{- $limits := $resources.limits | default dict }}
{{- $requests := $resources.requests | default dict }}
{{- if $limits.cpu }}
{{- $_ := set $resourceConfig "cpuLimit" $limits.cpu }}
{{- end }}
{{- if $requests.cpu }}
{{- $_ := set $resourceConfig "cpuRequest" $requests.cpu }}
{{- end }}
{{- if $limits.memory }}
{{- $_ := set $resourceConfig "memoryLimit" $limits.memory }}
{{- end }}
{{- if $requests.memory }}
{{- $_ := set $resourceConfig "memoryRequest" $requests.memory }}
{{- end }}
{
"resource": {
"cpuLimit": "{{ .Values.kserve.inferenceservice.resources.limits.cpu }}",
"cpuRequest": "{{ .Values.kserve.inferenceservice.resources.requests.cpu }}",
"memoryLimit": "{{ .Values.kserve.inferenceservice.resources.limits.memory }}",
"memoryRequest": "{{ .Values.kserve.inferenceservice.resources.requests.memory }}"
}
"resource": {{ $resourceConfig | toJson }}
}

opentelemetryCollector: |-
Expand Down
6 changes: 3 additions & 3 deletions charts/kserve-resources/templates/deployment.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -73,8 +73,8 @@ spec:
{{- toYaml . | nindent 10 }}
{{- end }}
resources:
{{- with .Values.kserve.controller.rbacProxy.resources }}
{{- toYaml . | nindent 10 }}
{{- if .Values.kserve.controller.rbacProxy.resources }}
{{- toYaml .Values.kserve.controller.rbacProxy.resources | indent 10 }}
{{- end }}
- command:
- /manager
Expand Down Expand Up @@ -113,7 +113,7 @@ spec:
resources:
{{- if .Values.kserve.controller.resources }}
{{ toYaml .Values.kserve.controller.resources | trim | indent 12 }}
{{- end }}
{{- end }}
ports:
- containerPort: 9443
name: webhook-server
Expand Down
25 changes: 4 additions & 21 deletions charts/kserve-resources/values.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -91,13 +91,7 @@ kserve:
# -- KServe controller manager rbac proxy contrainer image
rbacProxyImage: quay.io/brancz/kube-rbac-proxy:v0.18.0
rbacProxy:
resources:
limits:
cpu: 100m
memory: 300Mi
requests:
cpu: 100m
memory: 300Mi
resources: {}
securityContext:
allowPrivilegeEscalation: false
privileged: false
Expand Down Expand Up @@ -256,13 +250,7 @@ kserve:
# memory: 32Mi
#
# For more information, see [Resource Management for Pods and Containers](https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/).
resources:
limits:
cpu: 100m
memory: 300Mi
requests:
cpu: 100m
memory: 300Mi
resources: {}
# -- Indicates whether to create an addressable resolver ClusterRole for Knative Eventing.
# This ClusterRole grants the necessary permissions for the Knative's DomainMapping reconciler to resolve InferenceService addressables.
knativeAddressableResolver:
Expand Down Expand Up @@ -441,10 +429,5 @@ kserve:
security:
autoMountServiceAccountToken: true
inferenceservice:
resources:
limits:
cpu: "1"
memory: "2Gi"
requests:
cpu: "1"
memory: "2Gi"
# -- The default InferenceService resources limit.
resources: {}
0