vllm.v1.kv_offload.tiering ¶
Modules:
| Name | Description |
|---|---|
base | Abstract interfaces and data types for the secondary tiering layer. |
example | ExampleSecondaryTier: A simple in-memory secondary tier for testing. |
factory | Factory for creating secondary tier implementations. |
manager | TieringOffloadingManager: Multi-tier KV cache offloading orchestrator. |
spec | TieringOffloadingSpec: Spec for multi-tier KV cache offloading. |