OSDP: Optimal Sharded Data Parallel for Distributed Deep Learning
Abstract
Domain fit: AI-adjacent · Paper appears method- or tooling-adjacent to AI workflows with partial ecosystem coverage.
Large-scale deep learning models contribute to significant performance improvements on varieties of downstream tasks. Current data and model parallelism approaches utilize model replication and partition techniques to support the distributed training of ultra-large models. However, directly deploying these systems often leads to sub-optimal training efficiency due to the complex model architectures and the strict device memory constraints. In this paper, we propose Optimal Sharded Data Parallel (OSDP), an automated parallel training system that combines the advantages from both data and model parallelism. Given the model description and the device information, OSDP makes trade-offs between the memory consumption and the hardware utilization, thus automatically generates the distributed computation graph and maximizes the overall system throughput. In addition, OSDP introduces operator splitting to further alleviate peak memory footprints during training with negligible overheads, which enables the trainability of larger models as well as the higher throughput. Extensive experimental results of OSDP on multiple different kinds of large-scale models demonstrate that the proposed strategy outperforms the state-of-the-art in multiple regards.
Results and benchmarks
Large-scale deep learning models contribute to significant performance improvements on varieties of downstream tasks.
Benchmark evidence is limited
Evidence graph: 2 refs, 1 links.
Utility signals: depth 65/100, grounding 58/100, status medium.
Implementation
No direct implementation yet
Maintained implementation evidence is not confirmed for this paper yet.
Use the implementation status and reproduction sections for the current action plan.
No verified maintained repo yet
There is no verified maintained implementation yet. Use this baseline plan to decide whether to prototype now or defer.
- No direct maintained implementation was found. Use the paper PDF and citation graph to design a baseline reproduction.
- Start from related paper: On the Concepts of Parallelism in Biomolecular Computing.
- Start from this likely method family: Data modeling.
Time to first repro: a few days
Recommendation evidence is currently too limited for a maintained-repo choice. Use Implementation Status and Reproduction Path for a practical baseline plan.
- Estimate is based on paper-only reproduction flow
Reproduction readiness
No repo
No verified implementation available
- No maintained repository has been identified for this paper. Check adjacent implementations or HF artifacts below.
Hardware requirements
- Expect multi-day setup/compute for meaningful reproduction based on current guidance.
Validation caveat
Repositories and ecosystem
No additional verified repositories beyond the primary recommendation.
These repositories had low-confidence matching signals and are hidden by default.
- Youhe-Jiang/IJCAI2023-OptimalShardedDataParallel
Confidence: Low · 52 stars
Hugging Face artifacts
No trustworthy direct or curated related Hugging Face artifacts were found yet. Use targeted searches to quickly locate candidate models, datasets, and demos.
Models
Tip: start with models, then check datasets and spaces if you need evaluation data or demos.
Research context
7
Citations
33
References
Tasks
Computer science, Data parallelism, Computation, Throughput, Partition (number theory), Replication (statistics), Distributed memory, Distributed computing
Methods
Data modeling
Domains
Artificial intelligence, Computer Vision and Pattern Recognition
Related papers
- On the Concepts of Parallelism in Biomolecular ComputingSearch on Paper2Code
2018 · Semantic similarity
- On the duality between Or-parallelism and And-parallelism in logic programmingSearch on Paper2Code
1995 · Semantic similarity
- Relating data-parallelism and (and-) parallelism in logic programsSearch on Paper2Code
1995 · Semantic similarity
- Relating data-parallelism and (and-) parallelism in logic programsSearch on Paper2Code
1996 · Semantic similarity
- Intra-function ParallelismSearch on Paper2Code
2001 · Semantic similarity
- Parallelizing DNN Training on GPUs: Challenges and OpportunitiesSearch on Paper2Code
2021 · Semantic similarity
Open this paper in HFEPX to review benchmark signals, evaluation modes, and human-feedback protocol context.
Open in HFEPXJump to Paper2Code search queries derived from this paper's research context.