Metraipolis › archive › dc_f7ed950aff6c
Ordered Chaos: Towards Adaptive Multipath Load Balancing with Precise Fast Loss Detection — arXiv:2610.41707
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.41707
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T08:14:28.480Z
- SHA-256 of the archived text
1b865fb4f3661e0d903fc533d4e5ee3f096e94e4d51fb0d77cbfec7de54aac84- Stable URL of this page
https://metraipolis.com/documento/dc_f7ed950aff6c
How to cite
Ordered Chaos: Towards Adaptive Multipath Load Balancing with Precise Fast Loss Detection — arXiv:2610.41707 — arXiv (en), https://arxiv.org/abs/2610.41707 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f7ed950aff6c (Metraipolis archive dc_f7ed950aff6c, sha256 1b865fb4f3661e0d903fc533d4e5ee3f096e94e4d51fb0d77cbfec7de54aac84)
BibTeX
@misc{dcf7ed950aff6c,
title = {Ordered Chaos: Towards Adaptive Multipath Load Balancing with Precise Fast Loss Detection — arXiv:2610.41707},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.41707},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f7ed950aff6c; sha256 1b865fb4f3661e0d903fc533d4e5ee3f096e94e4d51fb0d77cbfec7de54aac84},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41707. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Load balancing in best-effort datacenter fabrics faces a tradeoff. Order-preserving schemes such as Equal-Cost Multipath (ECMP) retain loss semantics but use path diversity coarsely, while packet spraying balances load well but removes the ordering signals needed for fast, precise loss detection. We present Ordered Chaos, an end-host design that preserves enough structure for reliable loss detection without giving up fine-grained multipath routing. Reactive Subflow Spraying (RSS) divides each flow into ordered, dynamically routed subflows. Precise Fast Loss Detection (PFLD) combines this ordering with probes to distinguish reordering from congestion, corruption, tail, and retransmission losses within one RTT. Across simulations, AI and HPC workloads, and a DPU prototype, RSS+PFLD improves tail flow completion time by up to $3\times$, eliminates up to 99% of timeout-based recoveries, and accelerates collective and training communication by up to 35%. Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.NI, nuovi lavori del giorno). Categorie arXiv: Networking and Internet Architecture (cs.NI). Pagina del lavoro: https://arxiv.org/abs/2610.41707 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-03T08:03:08.887Z dalla pagina https://arxiv.org/list/cs.NI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).