vllm.distributed.kv_transfer.kv_connector.v1.mooncake.mooncake_utils ¶
EngineEntry dataclass ¶
Source code in vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_utils.py
MooncakeBootstrapServer ¶
A centralized server running on the global rank 0 prefiller worker. Prefiller workers register their connection info (IP, port, ranks) here.
Source code in vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_utils.py
35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 | |
__del__ ¶
__init__ ¶
__init__(vllm_config: VllmConfig, host: str, port: int)
Source code in vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_utils.py
_register_routes ¶
Source code in vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_utils.py
query async ¶
query() -> dict[int, EngineEntry]
register_worker async ¶
register_worker(payload: RegisterWorkerPayload)
Handles registration of a prefiller worker.
Source code in vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_utils.py
shutdown ¶
Source code in vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_utils.py
start ¶
Source code in vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_utils.py
RegisterWorkerPayload ¶
Bases: BaseModel