Pith. sign in

REVIEW 2 cited by

An Extensible Software Transport Layer for GPU Networking

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.17307 v2 pith:GZJDTBHQ submitted 2025-04-24 cs.NI

Yang Zhou , Zhongjie Chen , Ziming Mao , ChonLam Lao , Shuo Yang , Pravein Govindan Kannan , Jiaqi Gao , Yilong Zhao
show 6 more authors
Yongji Wu Kaichao You Fengyuan Ren Zhiying Xu Costin Raiciu Ion Stoica
This is my paper · ORCID
classification cs.NI
keywords transportrdmanetworkingnicssoftwareucclworkloadscollisions
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Fast-evolving machine learning (ML) workloads have increasing requirements for networking. However, host network transport on RDMA NICs is hard to evolve, causing problems for ML workloads. For example, single-path RDMA traffic is prone to flow collisions that severely degrade collective communication performance. We present UCCL, an extensible software transport layer to evolve GPU networking. UCCL decouples the data path and control path of existing RDMA NICs and efficiently runs the control-path transport on host CPUs. This software extensibility brings in transport innovations that cannot be achieved in hardware for ML workloads, e.g., a multipath transport to resolve flow collisions. ML collectives atop UCCL achieve up to 4.5x higher performance compared to existing RDMA NICs.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. OrderMoE: An expert similarity driven distributed edge MoE inference

    cs.NI 2026-07 conditional novelty 6.0 of 10

    OrderMoE reduces cross-server traffic in edge MoE inference by substituting a similar local expert for the router's remote target expert, with only small quality loss.

  2. TENT: A Declarative Slice Spraying Engine for Performant and Resilient Data Movement in Disaggregated LLM Serving

    cs.DC 2026-04 conditional novelty 6.0 of 10

    A telemetry-driven slice-spraying transfer engine with late-binding path selection improves LLM serving throughput and self-heals from link failures in tens of milliseconds.

Pith tools