paper-with-me

Papers

Placement Semantics for Distributed Deep Learning: A Systematic Framework for Analyzing Parallelism Strategies

2026-01-05 · Deep Pankajbhai Mehta arxiv

Training large language models requires distributing computation across many accelerators, yet practitioners select parallelism strategies (data, tensor, pipeline, ZeRO) through trial and error because no unified systematic framework predicts their behavior. We introduce placement semantics: each strategy is specified by how it places four training states (parameters, optimizer, gradients, activations) across devices using five modes (replicated, sharded, sharded-with-gather, materialized, offloaded). From placement alone, without implementation details, we derive memory consumption and communication volume. Our predictions match published results exactly: ZeRO-3 uses 8x less memory than data parallelism at 1.5x communication cost, as reported in the original paper. We prove two conditions (gradient integrity, state consistency) are necessary and sufficient for distributed training to match single-device results, and provide composition rules for combining strategies safely. The framework unifies ZeRO Stages 1-3, Fully Sharded Data Parallel (FSDP), tensor parallelism, and pipeline parallelism as instances with different placement choices.

📄 PDF Abstract BibTeX arXiv:2601.02311

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Towards a Distributed Federated Learning Aggregation Placement using Particle Swarm Intelligence

2025-04-22 · Amir Ali-Pour, Sadra Bekrani, Laya Samizadeh, Julien Gascon-Samson

Federated learning has become a promising distributed learning concept with extra insurance on data privacy. Extensive studies on various models of Federated learning have been done since the coinage of its term. One of …

Federated Learning

Interpreting vision transformers via residual replacement model

2025-09-22 · Jinyeong Kim, Junhyeok Kim, Yumin Shim, Joohyeok Kim 외 arxiv

How do vision transformers (ViTs) represent and process the world? This paper addresses this long-standing question through the first systematic analysis of 6.6K features across all layers, extracted via sparse autoencod…

End-to-end Adaptive Distributed Training on PaddlePaddle

2021-12-06 · Yulong Ao, Zhihua Wu, dianhai yu, Weibao Gong 외

Distributed training has become a pervasive and effective approach for training a large neural network (NN) model with processing massive data. However, it is very challenging to satisfy requirements from various NN mode…

CPUGPULanguage ModellingRecommendation Systems+1

FleetX

2021-12-06 · arXiv 2021 12 · EW

Distributed training has become a pervasive and effective approach for training a large neural network (NN) model with processing massive data. However, it is very challenging to satisfy requirements from various NN mo…

CPUDistributed ComputingGPULanguage Modelling+2

Semantics, Representations and Grammars for Deep Learning

2015-09-29 · David Balduzzi

Deep learning is currently the subject of intensive study. However, fundamental concepts such as representations are not formally defined -- researchers "know them when they see them" -- and there is no common language f…

Deep LearningDistributed Optimization