general family · filed under optimization
Pipeline Parallelism
A model-parallelism family that divides model execution into pipeline stages; the evidence does not specify a more particular scheduling method.
- sources
- 2
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
optional 1used 1
Documented in
Evidence
2 spans quoted from the sources, strongest treatment first.
2-way pipeline parallelism
usedsoftware implementationin Nemotron 3 UltraNVIDIA
--pipeline-parallel-size 2
optionalinference servingin Nemotron 3 UltraNVIDIA
Filed alongside
Other methods under optimization :: training parallelism.
Expert ParallelismTensor ParallelismMoonEPCommunication-Computation OverlapAll-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 ZeRO-2 Gradient Sharding with CPU Offloadingpipeline-bubble scheduling of ViT computationRedundant-Expert Capacity ReservationSConv-Aware Tensor-Parallel ShardingSequence Parallelism for Activations