Support gated launch to defer startup memory allocation (#35927)
This commit is contained in:
@@ -1046,6 +1046,11 @@ class ServerArgs:
|
||||
),
|
||||
NS("parallel"),
|
||||
] = None
|
||||
gated_launch_port: A[
|
||||
Optional[int],
|
||||
"The port of the gated launch control server. When set, every rank blocks right after the distributed environment is initialized, before any sizable GPU allocation, until `POST /gate/activate` is sent to this port on the host of the first rank. This lets an external orchestrator defer the memory hungry part of startup to a safe window. Defaults to None, which disables the gate.",
|
||||
NS("parallel"),
|
||||
] = None
|
||||
nnodes: A[int, "The number of nodes.", NS("parallel")] = 1
|
||||
node_rank: A[int, "The node rank.", NS("parallel")] = 0
|
||||
tp_size: A[
|
||||
|
||||
Reference in New Issue
Block a user