### What this PR does / why we need it?
Add DeepSeek-V3.2-W8A8 nightly ci test:
DeepSeek-V3.2-W8A8 1node DP2+TP8
:tests/e2e/nightly/models/test_deepseek_v3_2_w8a8.py
### Does this PR introduce _any_ user-facing change
- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c
Signed-off-by: hfadzxy <starmoon_zhang@163.com>
126 lines
3.9 KiB
Django/Jinja
126 lines
3.9 KiB
Django/Jinja
apiVersion: leaderworkerset.x-k8s.io/v1
|
|
kind: LeaderWorkerSet
|
|
metadata:
|
|
name: vllm
|
|
namespace: vllm-project
|
|
spec:
|
|
replicas: {{ replicas | default(1) }}
|
|
leaderWorkerTemplate:
|
|
size: {{ size | default(2) }}
|
|
restartPolicy: None
|
|
leaderTemplate:
|
|
metadata:
|
|
labels:
|
|
role: leader
|
|
spec:
|
|
containers:
|
|
- name: vllm-leader
|
|
imagePullPolicy: Always
|
|
image: {{ image | default("swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/vllm-ascend:nightly-a3") }}
|
|
env:
|
|
- name: CONFIG_YAML_PATH
|
|
value: {{ config_file_path | default("DeepSeek-V3.yaml") }}
|
|
- name: WORKSPACE
|
|
value: "/vllm-workspace"
|
|
- name: FAIL_TAG
|
|
value: {{ fail_tag | default("FAIL_TAG") }}
|
|
command:
|
|
- sh
|
|
- -c
|
|
- |
|
|
bash /root/.cache/tests/run.sh
|
|
resources:
|
|
limits:
|
|
huawei.com/ascend-1980: {{ npu_per_node | default("16") }}
|
|
memory: 512Gi
|
|
ephemeral-storage: 100Gi
|
|
requests:
|
|
huawei.com/ascend-1980: {{ npu_per_node | default("16") }}
|
|
ephemeral-storage: 100Gi
|
|
cpu: 125
|
|
ports:
|
|
- containerPort: 8080
|
|
# readinessProbe:
|
|
# tcpSocket:
|
|
# port: 8080
|
|
# initialDelaySeconds: 15
|
|
# periodSeconds: 10
|
|
volumeMounts:
|
|
- mountPath: /root/.cache
|
|
name: shared-volume
|
|
- mountPath: /usr/local/Ascend/driver/tools
|
|
name: driver-tools
|
|
- mountPath: /dev/shm
|
|
name: dshm
|
|
volumes:
|
|
- name: dshm
|
|
emptyDir:
|
|
medium: Memory
|
|
sizeLimit: 15Gi
|
|
- name: shared-volume
|
|
persistentVolumeClaim:
|
|
claimName: nv-action-vllm-benchmarks-v2
|
|
- name: driver-tools
|
|
hostPath:
|
|
path: /usr/local/Ascend/driver/tools
|
|
workerTemplate:
|
|
spec:
|
|
containers:
|
|
- name: vllm-worker
|
|
imagePullPolicy: Always
|
|
image: {{ image | default("swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/vllm-ascend:nightly-a3") }}
|
|
env:
|
|
- name: CONFIG_YAML_PATH
|
|
value: {{ config_file_path | default("DeepSeek-V3.yaml") }}
|
|
- name: WORKSPACE
|
|
value: "/vllm-workspace"
|
|
- name: FAIL_TAG
|
|
value: {{ fail_tag | default("FAIL_TAG") }}
|
|
command:
|
|
- sh
|
|
- -c
|
|
- |
|
|
bash /root/.cache/tests/run.sh
|
|
resources:
|
|
limits:
|
|
huawei.com/ascend-1980: {{ npu_per_node | default("16") }}
|
|
memory: 512Gi
|
|
ephemeral-storage: 100Gi
|
|
requests:
|
|
huawei.com/ascend-1980: {{ npu_per_node | default("16") }}
|
|
ephemeral-storage: 100Gi
|
|
cpu: 125
|
|
volumeMounts:
|
|
- mountPath: /root/.cache
|
|
name: shared-volume
|
|
- mountPath: /usr/local/Ascend/driver/tools
|
|
name: driver-tools
|
|
- mountPath: /dev/shm
|
|
name: dshm
|
|
volumes:
|
|
- name: dshm
|
|
emptyDir:
|
|
medium: Memory
|
|
sizeLimit: 15Gi
|
|
- name: shared-volume
|
|
persistentVolumeClaim:
|
|
claimName: nv-action-vllm-benchmarks-v2
|
|
- name: driver-tools
|
|
hostPath:
|
|
path: /usr/local/Ascend/driver/tools
|
|
---
|
|
apiVersion: v1
|
|
kind: Service
|
|
metadata:
|
|
name: vllm-leader
|
|
namespace: vllm-project
|
|
spec:
|
|
ports:
|
|
- name: http
|
|
port: 8080
|
|
protocol: TCP
|
|
targetPort: 8080
|
|
selector:
|
|
leaderworkerset.sigs.k8s.io/name: vllm
|
|
role: leader
|
|
type: ClusterIP |