Update Helm chart

[jjk-g]

Changes helm chart to expose the full config via values.yaml. Helm chart usage can overwrite the values with a custom yaml file or --set argument of the install command.

Example usage test.yaml:

config:
  load:
    type: constant
    stages:
    - rate: 2
      duration: 10
  server:
    base_url: http://llama3-8b-vllm-service:8000
  data:
    type: shareGPT
    input_distribution:
      max: 1024
    output_distribution:
      max: 1024

install output:

$ helm install test . --dry-run -f test.
yaml 
NAME: test
LAST DEPLOYED: Thu Jul 24 04:20:18 2025
NAMESPACE: default
STATUS: pending-install
REVISION: 1
TEST SUITE: None
HOOKS:
MANIFEST:
---
# Source: inference-perf/templates/configmap.yaml
# inference-perf/templates/configmap.yaml
apiVersion: v1
kind: ConfigMap
metadata:
  name: test-inference-perf-config
  labels:
    helm.sh/chart: inference-perf-0.1.0
    app.kubernetes.io/name: inference-perf
    app.kubernetes.io/instance: test
    app.kubernetes.io/version: "0.1.0"
    app.kubernetes.io/managed-by: Helm
data:
  config.yml: |
    api:
      type: completion
    data:
      input_distribution:
        max: 1024
      output_distribution:
        max: 1024
      type: shareGPT
    load:
      stages:
      - duration: 10
        rate: 2
      type: constant
    metrics:
      prometheus:
        scrape_interval: 15
        url: http://localhost:9090
      type: prometheus
    report:
      prometheus:
        per_stage: false
        summary: true
      request_lifecycle:
        per_request: false
        per_stage: true
        summary: true
    server:
      base_url: http://llama3-8b-vllm-service:8000
      ignore_eos: true
      type: vllm
---
# Source: inference-perf/templates/job.yaml
# inference-perf/templates/job.yaml
apiVersion: batch/v1
kind: Job
metadata:
  name: test-inference-perf-job
  labels:
    helm.sh/chart: inference-perf-0.1.0
    app.kubernetes.io/name: inference-perf
    app.kubernetes.io/instance: test
    app.kubernetes.io/version: "0.1.0"
    app.kubernetes.io/managed-by: Helm
    app: inference-perf
spec:
  template:
    metadata:
      labels:
        app.kubernetes.io/name: inference-perf
        app.kubernetes.io/instance: test
        app: inference-perf
    spec:
      restartPolicy: Never
      containers:
        - name: inference-perf-container
          image: 
          command: ["inference-perf"]
          args: 
            - "--config_file"
            - "config.yml"
            - "--log-level"
            - INFO
          env:
          volumeMounts:
            - name: config-volume
              mountPath: /cfg
              readOnly: true
          resources:
            requests:
              memory: 8G
      volumes:
        - name: config-volume
          configMap:
            name: test-inference-perf-config
---
# Source: inference-perf/templates/secret.yaml
# inference-perf/templates/secret.yaml

[achandrasekar]

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: achandrasekar, jjk-g

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details

Needs approval from an approver in each of these files:

• OWNERS [achandrasekar]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment