llm-d-inference-scheduler: Network¶
Service Map¶
graph LR
classDef svc fill:#2ecc71,stroke:#27ae60,color:#fff
classDef test fill:#95a5a6,stroke:#7f8c8d,color:#fff
classDef component fill:#3498db,stroke:#2980b9,color:#fff
classDef ext fill:#e74c3c,stroke:#c0392b,color:#fff
llm_d_inference_scheduler["llm-d-inference-scheduler"]:::component
llm_d_inference_scheduler --> svc_0["${EPP_NAME}\nClusterIP: 5557/TCP,9002/TCP,9090/TCP"]:::svc
llm_d_inference_scheduler --> svc_1["e2e-epp\nClusterIP: 5557/TCP,9002/TCP"]:::test
llm_d_inference_scheduler --> svc_2["e2e-epp-health\nNodePort: 9003/TCP"]:::test
llm_d_inference_scheduler --> svc_3["e2e-epp-metrics\nNodePort: 9090/TCP"]:::test
llm_d_inference_scheduler --> svc_4["inference-gateway-istio-nodeport\nNodePort: 15021/TCP,80/TCP"]:::svc
llm_d_inference_scheduler --> svc_5["istiod-llm-d-gateway\nClusterIP: 15010/TCP,15012/TCP,15014/TCP,443/TCP"]:::svc
llm_d_inference_scheduler --> svc_6["service\nClusterIP: 8080/TCP"]:::svc
llm_d_inference_scheduler -.-> ext_grpc[["grpc\ngrpc"]]:::ext
Services¶
| Name | Type | Ports | Source |
|---|---|---|---|
| ${EPP_NAME} | ClusterIP | 9002/TCP, 5557/TCP, 9090/TCP | deploy/components/inference-gateway/services.yaml |
| e2e-epp | ClusterIP | 9002/TCP, 5557/TCP | test/e2e/yaml/services.yaml |
| e2e-epp-health | NodePort | 9003/TCP | test/e2e/yaml/services.yaml |
| e2e-epp-metrics | NodePort | 9090/TCP | test/e2e/yaml/services.yaml |
| inference-gateway-istio-nodeport | NodePort | 15021/TCP, 80/TCP | deploy/environments/dev/base-kind-istio/services.yaml |
| istiod-llm-d-gateway | ClusterIP | 15010/TCP, 15012/TCP, 443/TCP, 15014/TCP | deploy/components/istio-control-plane/services.yaml |
| service | ClusterIP | 8080/TCP | deploy/environments/kubernetes-base/common/service.yaml |
Ingress / Routing¶
| Kind | Name | Hosts | Paths | TLS | Source |
|---|---|---|---|---|---|
| Gateway | inference-gateway | no | deploy/components/inference-gateway/gateways.yaml |
||
| Gateway | inference-gateway | no | test/sidecar/config/gateway/gateway.yaml |
||
| HTTPRoute | ${POOL_NAME}-inference-route | / | no | deploy/components/inference-gateway/httproutes.yaml |
|
| Route | route | yes | deploy/environments/kubernetes-base/openshift/route.yaml |
No Network Policies
No NetworkPolicy resources found. All pod-to-pod traffic is allowed by default.