58 lines
1.0 KiB
YAML
58 lines
1.0 KiB
YAML
# Copyright (C) 2024 Intel Corporation
|
|
# SPDX-License-Identifier: Apache-2.0
|
|
|
|
HUGGINGFACEHUB_API_TOKEN: ${HF_TOKEN}
|
|
|
|
podSpecs:
|
|
- name: chatqna-backend-server-deploy
|
|
spec:
|
|
replicas: 2
|
|
resources:
|
|
limits:
|
|
cpu: "8"
|
|
memory: "8000Mi"
|
|
requests:
|
|
cpu: "8"
|
|
memory: "8000Mi"
|
|
|
|
- name: embedding-dependency-deploy
|
|
spec:
|
|
replicas: 1
|
|
resources:
|
|
limits:
|
|
cpu: "80"
|
|
memory: "20000Mi"
|
|
requests:
|
|
cpu: "80"
|
|
memory: "20000Mi"
|
|
|
|
- name: reranking-dependency-deploy
|
|
spec:
|
|
replicas: 1
|
|
resources:
|
|
limits:
|
|
habana.ai/gaudi: 1
|
|
|
|
- name: llm-dependency-deploy
|
|
spec:
|
|
replicas: 7
|
|
resources:
|
|
limits:
|
|
habana.ai/gaudi: 1
|
|
|
|
- name: dataprep-deploy
|
|
spec:
|
|
replicas: 1
|
|
|
|
- name: vector-db
|
|
spec:
|
|
replicas: 1
|
|
|
|
- name: retriever-deploy
|
|
spec:
|
|
replicas: 2
|
|
resources:
|
|
requests:
|
|
cpu: "4"
|
|
memory: "4000Mi"
|