Papers › DSP: Dynamic Sequence Parallelism for Multi-Dimensional Transformers

DSP: Dynamic Sequence Parallelism for Multi-Dimensional Transformers

15 Mar 2024arXiv:2403.10266archive 2025-07-28

Xuanlei Zhao, Shenggan Cheng, Chang Chen, Zangwei Zheng, Ziming Liu, Zheming Yang, Yang You

Scaling multi-dimensional transformers to long sequences is indispensable across various domains. However, the challenges of large memory requirements and slow speeds of such sequences necessitate sequence parallelism. All existing approaches fall under the category of embedded sequence parallelism, which are limited to shard along a single sequence dimension, thereby introducing significant communication overhead. However, the nature of multi-dimensional transformers involves independent calculations across multiple sequence dimensions. To this end, we propose Dynamic Sequence Parallelism (DSP) as a novel abstraction of sequence parallelism. DSP dynamically switches the parallel dimension among all sequences according to the computation stage with efficient resharding strategy. DSP offers significant reductions in communication costs, adaptability across modules, and ease of implementation with minimal constraints. Experimental evaluations demonstrate DSP's superiority over state-of-the-art embedded sequence parallelism methods by remarkable throughput improvements ranging from 32.2% to 10x, with less than 25% communication volume.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2403.10266")

Code

Syntology Ran 12 of 19 code samples harvested from 2 repositories linked to this paper; 7 have no recorded run. Of those that ran: 1 ran · honoured contract; 1 ran · violated contract; 2 ran · our draft was wrong; 8 ran with no contract checked.

By repository: official repository: 18 samples from 1 repository, 12 ran; community (archive-listed): 1 sample from 1 repository, 0 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

nus-hpc-ai-lab/opendit officialmentioned in papermentioned on GitHubpytorchApache-2.0 report
NUS-HPC-AI-Lab/VideoSys mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

19 samples harvested; 12 ran; 1 honoured the contract we drafted; 7 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · honoured contract
1ran · violated contract
2ran · our draft was wrong
8ran
7unverified

Licence: 0 of the 19 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; each sample names its own and says which. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

Normalize nus-hpc-ai-lab/opendit/videosys/models/autoencoders/autoencoder_kl_open_sora_plan_v110.py official repository ran · our draft was wrong Apache-2.0 (permissive) · 9fcdaa6e423e8aa7 · report
all_to_all_comm nus-hpc-ai-lab/opendit/videosys/core/distributed/comm.py official repository ran Apache-2.0 (permissive) · 32447bb9476ba9a4 · report
cast_tuple nus-hpc-ai-lab/opendit/videosys/models/autoencoders/autoencoder_kl_open_sora.py official repository ran · honoured contract fingerprinted Apache-2.0 (permissive) · c1b68cdd9e73316a · report
divisible_by nus-hpc-ai-lab/opendit/videosys/models/autoencoders/autoencoder_kl_open_sora.py official repository ran · violated contract fingerprinted Apache-2.0 (permissive) · f912c209e9c899ca · report
get_distributed_init_method nus-hpc-ai-lab/opendit/videosys/core/engine/mp_utils.py official repository ran fingerprinted Apache-2.0 (permissive) · 6875a0fdda4fa123 · report
if_broadcast_cross nus-hpc-ai-lab/opendit/videosys/core/pab/pab_mgr.py official repository ran Apache-2.0 (permissive) · daeff11d0e2563e3 · report
if_broadcast_spatial nus-hpc-ai-lab/opendit/videosys/core/pab/pab_mgr.py official repository ran Apache-2.0 (permissive) · 90d456a8c999fe52 · report
if_broadcast_temporal nus-hpc-ai-lab/opendit/videosys/core/pab/pab_mgr.py official repository ran Apache-2.0 (permissive) · b1810f6a3b221de3 · report
nonlinearity nus-hpc-ai-lab/opendit/videosys/models/autoencoders/autoencoder_kl_open_sora_plan_v110.py official repository ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 3137073275f8c21a · report
set_distributed_state nus-hpc-ai-lab/opendit/videosys/core/distributed/parallel_mgr.py official repository ran Apache-2.0 (permissive) · 12c9db8fb79f543d · report
split_sequence nus-hpc-ai-lab/opendit/videosys/core/distributed/comm.py official repository ran Apache-2.0 (permissive) · 4809422d25cbda07 · report
tensor_to_video nus-hpc-ai-lab/opendit/videosys/models/autoencoders/autoencoder_kl_open_sora_plan_v110.py official repository ran Apache-2.0 (permissive) · 5d0218d6974725ed · report
add_timestamp nus-hpc-ai-lab/opendit/videosys/core/dcp/recompute.py official repository unverified Apache-2.0 (permissive) · 686ce42a4e575400 · report
auto_recompute nus-hpc-ai-lab/opendit/videosys/core/dcp/recompute.py official repository unverified Apache-2.0 (permissive) · 17f768f6ac1bbb38 · report
gather_sequence nus-hpc-ai-lab/opendit/videosys/core/distributed/comm.py official repository unverified Apache-2.0 (permissive) · 40ebfee15a7cbaa6 · report
generate nus-hpc-ai-lab/opendit/gradio/cogvideox.py official repository unverified Apache-2.0 (permissive) · be3706734550c9d9 · report
is_odd nus-hpc-ai-lab/opendit/videosys/models/autoencoders/autoencoder_kl_open_sora.py official repository unverified Apache-2.0 (permissive) · 945442cc2f6a19c9 · report
recompute_func nus-hpc-ai-lab/opendit/videosys/core/dcp/recompute.py official repository unverified Apache-2.0 (permissive) · 73301f9777f1a2c0 · report
DynamicParallelManager NUS-HPC-AI-Lab/VideoSys/videosys/core/distributed/parallel_mgr.py community (archive-listed) unverified Apache-2.0 (permissive) · 24afe9113375d56e · report

Tasks

Text GenerationVideo Generation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections