specific method · filed under optimization
Pipeline ZeRO-2 Gradient Sharding with CPU Offloading
A pipeline training method that shards gradients across data-parallel ranks and stores the shards in CPU memory.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
we use Pipeline ZeRO-2 gradient sharding [145] to shard gradients across data-parallel (DP) ranks. Furthermore, we store the sharded gradients in CPU memory
usedoptimizationin Kimi K3Moonshot AI
Filed alongside
Other methods under optimization :: training parallelism.
Expert ParallelismTensor ParallelismMoonEPCommunication-Computation OverlapPipeline ParallelismAll-to-all Gradient Exchange with Local FP32 SummationCache-Based Pipeline CommunicationContext ParallelismData Replica Reduction over the Data Center NetworkData-Weighted Data ParallelismDynamic Context Parallelism for Large Multimodal SamplesFully Balanced Expert-Parallel TrainingGPU Planning Kernel for Redundant Expert MigrationHybrid ZeRO Bucket Assignment for MuonKDA Context ParallelismKnapsack-Based Balanced Assignment of Dense Parameter MatricesLoad-Balanced Image ShardingModified DualPipe 1F1B Pipeline Overlap for mHCNS-FLOP-Balanced Static Parameter PartitioningPipeline Payload Extensionspipeline-bubble scheduling of ViT computationRedundant-Expert Capacity ReservationSConv-Aware Tensor-Parallel ShardingSequence Parallelism for Activations