mscclpp.language.default_algos
- mscclpp.language.default_algos.allreduce_2nodes(spec, thread_block_group_size)
Implements a multi-node AllReduce using a hierarchical approach: 1. Intra-node allreduce 2. Inter-node exchange (exchange reduced data between nodes) 3. Intra-node allreduce
- Return type:
- Parameters:
Modules
|
Implements a multi-node AllReduce using a hierarchical approach: 1. |