Skip to content

[build]: Updating vllm deployment to support the latest images and scorers. #112

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.

Sign up for GitHub

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

Merged
merged 1 commit into from
May 5, 2025
Merged

[build]: Updating vllm deployment to support the latest images and scorers. #112

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
9 changes: 6 additions & 3 deletions deploy/components/vllm-p2p/vllm-deployment.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,13 +31,12 @@ spec:
- "-c"
args:
- |
export LMCACHE_DISTRIBUTED_URL=$${${POD_IP}}:80 && \
export LMCACHE_DISTRIBUTED_URL=$${${POD_IP}} && \
vllm serve ${MODEL_NAME} \
--host 0.0.0.0 \
--port 8000 \
--enable-chunked-prefill false \
--max-model-len ${MAX_MODEL_LEN} \
--kv-transfer-config '{"kv_connector":"LMCacheConnector","kv_role":"kv_both"}'
--kv-transfer-config '{"kv_connector":"LMCacheConnectorV1","kv_role":"kv_both"}'
ports:
- name: http
containerPort: 8000
Expand Down Expand Up @@ -78,6 +77,10 @@ spec:
secretKeyRef:
name: ${HF_SECRET_NAME}
key: ${HF_SECRET_KEY}
- name: VLLM_ENABLE_V1_MULTIPROCESSING
value: "1"
- name: VLLM_WORKER_MULTIPROC_METHOD
value: spawn
- name: LMCACHE_LOOKUP_URL
value: ${REDIS_HOST}:${REDIS_PORT}
- name: LMCACHE_ENABLE_DEBUG
Expand Down
10 changes: 9 additions & 1 deletion deploy/environments/dev/kubernetes-kgateway/patch-deployments.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,4 +29,12 @@ spec:
valueFrom:
secretKeyRef:
name: hf-token
key: ${HF_SECRET_KEY}
key: ${HF_SECRET_KEY}
- name: ENABLE_KVCACHE_AWARE_SCORER
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

we'll need to find a more elegant way of loading/enabling specific plugins (more than just filters and/or scorers) - but ok for now.

value: "true"
- name: KVCACHE_AWARE_SCORER_WEIGHT
value: "2.0"
- name: ENABLE_LOAD_AWARE_SCORER
value: "true"
- name: LOAD_AWARE_SCORER_WEIGHT
value: "1.0"
8 changes: 4 additions & 4 deletions scripts/kubernetes-dev-env.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,10 +65,10 @@ case "${VLLM_MODE}" in
export LORA_ADAPTER_SYNCER_TAG="${LORA_ADAPTER_SYNCER_TAG:-v20250425-ddc3d69}"

elif [[ "$VLLM_MODE" == "vllm-p2p" ]]; then
export VLLM_IMAGE="${VLLM_IMAGE:-lmcache/vllm-openai}"
export VLLM_TAG="${VLLM_TAG:-2025-03-10}"
export EPP_IMAGE="${EPP_IMAGE:- quay.io/vmaroon/gateway-api-inference-extension/epp}"
export EPP_TAG="${EPP_TAG:-kv-aware}"
export VLLM_IMAGE="${VLLM_IMAGE:-quay.io/llm-d/llm-d-dev}"
export VLLM_TAG="${VLLM_TAG:-lmcache-0.0.6-amd64}"
export EPP_IMAGE="${EPP_IMAGE:-quay.io/llm-d/llm-d-gateway-api-inference-extension-dev}"
export EPP_TAG="${EPP_TAG:-0.0.5-amd64}"
export MAX_MODEL_LEN="${MAX_MODEL_LEN:-32768}"
export PVC_NAME="${PVC_NAME:-vllm-p2p-storage-claim}"
export PVC_ACCESS_MODE="${PVC_ACCESS_MODE:-ReadWriteOnce}"
Expand Down