Papers › Comet: Fine-grained Computation-communication Overlapping for Mixture-of-Experts

Comet: Fine-grained Computation-communication Overlapping for Mixture-of-Experts

27 Feb 2025arXiv:2502.19811archive 2025-07-28

Shulai Zhang, Ningxin Zheng, Haibin Lin, Ziheng Jiang, Wenlei Bao, Chengquan Jiang, Qi Hou, Weihao Cui, Size Zheng, Li-Wen Chang, Quan Chen, Xin Liu

Mixture-of-experts (MoE) has been extensively employed to scale large language models to trillion-plus parameters while maintaining a fixed computational cost. The development of large MoE models in the distributed scenario encounters the problem of large communication overhead. The inter-device communication of a MoE layer can occupy 47% time of the entire model execution with popular models and frameworks. Therefore, existing methods suggest the communication in a MoE layer to be pipelined with the computation for overlapping. However, these coarse grained overlapping schemes introduce a notable impairment of computational efficiency and the latency concealing is sub-optimal. To this end, we present COMET, an optimized MoE system with fine-grained communication-computation overlapping. Leveraging data dependency analysis and task rescheduling, COMET achieves precise fine-grained overlapping of communication and computation. Through adaptive workload assignment, COMET effectively eliminates fine-grained communication bottlenecks and enhances its adaptability across various scenarios. Our evaluation shows that COMET accelerates the execution of a single MoE layer by 1.96× and for end-to-end execution, COMET delivers a 1.71× speedup on average. COMET has been adopted in the production environment of clusters with ten-thousand-scale of GPUs, achieving savings of millions of GPU hours.

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="2502.19811")

Code

Syntology Ran 1 of 20 code samples harvested from 2 repositories linked to this paper; 19 have no recorded run. Of those that ran: 1 ran with no contract checked.

By repository: official repository: 17 samples from 1 repository, 0 ran; community (archive-listed): 3 samples from 1 repository, 1 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

bytedance/flux officialmentioned in papermentioned on GitHubpytorchApache-2.0 report
multi-swe-bench/magentless mentioned on GitHubMIT report
multi-swe-bench/mopenhands mentioned on GitHubMIT report
multi-swe-bench/mswe-agent mentioned on GitHub 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

20 samples harvested; 1 ran; 0 honoured the contract we drafted; 19 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
19unverified

Licence: 0 of the 20 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.

bench_func bytedance/flux/python/flux/util.py official repository unverified Apache-2.0 (permissive) · b2e2521687ac95ce · report
calc_gather_index bytedance/flux/python/flux/testing/moe_utils.py official repository unverified Apache-2.0 (permissive) · b0bd0e05915bf19c · report
calc_gather_index_stable bytedance/flux/python/flux/testing/moe_utils.py official repository unverified Apache-2.0 (permissive) · c7158b2a6c069b16 · report
calc_scatter_index_stable bytedance/flux/python/flux/testing/moe_utils.py official repository unverified Apache-2.0 (permissive) · c252afed940d6e30 · report
get_device_property bytedance/flux/python/flux/testing/gpu_perf_model.py official repository unverified Apache-2.0 (permissive) · bdad45577321a1d2 · report
get_dist_env bytedance/flux/python/flux/dist_utils.py official repository unverified Apache-2.0 (permissive) · c23cadd88b362499 · report
get_inter_node_ag_group bytedance/flux/python/flux/ag_kernel_internode.py official repository unverified Apache-2.0 (permissive) · 3e48284521d53051 · report
get_intra_node_ag_group bytedance/flux/python/flux/ag_kernel_internode.py official repository unverified Apache-2.0 (permissive) · 80683aed2fa79e07 · report
get_intra_node_pg_group bytedance/flux/python/flux/gemm_rs_sm80.py official repository unverified Apache-2.0 (permissive) · 673d58d820da47c9 · report
get_torch_prof_ctx bytedance/flux/python/flux/util.py official repository unverified Apache-2.0 (permissive) · 318a824a192ac02f · report
is_fp8_dtype bytedance/flux/python/flux/testing/gpu_perf_model.py official repository unverified Apache-2.0 (permissive) · f47e5cdd7171b461 · report
is_fp8_dtype bytedance/flux/python/flux/util.py official repository unverified Apache-2.0 (permissive) · 6469bfc20e491271 · report
nvsmi bytedance/flux/python/flux/testing/gpu_perf_model.py official repository unverified Apache-2.0 (permissive) · 0c0fc4a132ede3cb · report
run_perf bytedance/flux/python/flux/testing/utils.py official repository unverified Apache-2.0 (permissive) · 146ee989a1ad16cb · report
torch_post_attn_all_to_all_transpose bytedance/flux/python/flux/testing/ulysses_sp_utils.py official repository unverified Apache-2.0 (permissive) · 24ef3f53f604fc80 · report
torch_pre_attn_all_to_all_transpose bytedance/flux/python/flux/testing/ulysses_sp_utils.py official repository unverified Apache-2.0 (permissive) · 36283518ba9f282d · report
torch_pre_attn_qkv_pack_a2a bytedance/flux/python/flux/testing/ulysses_sp_utils.py official repository unverified Apache-2.0 (permissive) · e693c5491138e1e3 · report
construct_file_meta_data multi-swe-bench/magentless/agentless/fl/Index.py community (archive-listed) ran MIT (permissive) · 4dc6308a692bee4f · report
parse_patch multi-swe-bench/magentless/get_repo_structure/get_patch_info.py community (archive-listed) unverified MIT (permissive) · f2c4e9391d254c7e · report
parse_python_file multi-swe-bench/magentless/get_repo_structure/get_repo_structure.py community (archive-listed) unverified MIT (permissive) · 94dc3cbe589f0afa · report

Tasks

Computational EfficiencyMixture-of-Experts

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

MoE

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