Skip to content

Commit f2f1dfd

Browse files
committed
changes in llm-d pr 548
Signed-off-by: Michael Kalantar <kalantar@us.ibm.com>
1 parent 6ad14b6 commit f2f1dfd

2 files changed

Lines changed: 11 additions & 4 deletions

File tree

scenarios/guides/wide-ep-lws.sh

Lines changed: 0 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -60,9 +60,6 @@ cat << EOF > $LLMDBENCH_VLLM_MODELSERVICE_INFERENCE_POOL_PROVIDER_CONFIG
6060
destinationRule:
6161
host: REPLACE_ENV_LLMDBENCH_DEPLOY_CURRENT_MODEL_ID_LABEL-gaie-epp
6262
trafficPolicy:
63-
tls:
64-
mode: SIMPLE
65-
insecureSkipVerify: true
6663
connectionPool:
6764
http:
6865
http1MaxPendingRequests: 256000

setup/steps/07_deploy_setup.py

Lines changed: 11 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -19,8 +19,18 @@ def gateway_values(provider : str, host: str, service: str) -> str:
1919
gatewayClassName: istio
2020
gatewayParameters:
2121
enabled: true
22+
accessLogging: false
23+
logLevel: error
24+
resources:
25+
limits:
26+
cpu: "16"
27+
memory: 16Gi
28+
requests:
29+
cpu: "4"
30+
memory: 4Gi
2231
service:
23-
type: {service}"""
32+
type: {service}
33+
"""
2434

2535
elif provider == "kgateway":
2636
return f"""gateway:

0 commit comments

Comments
 (0)