Metraipolis › archive › dc_671d3f55cd3d
TideRL: Boosting Agentic RL Goodput with Readiness-Aware Scheduling — arXiv:2608.10402
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.10402
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T06:14:38.382Z
- SHA-256 of the archived text
6907619d6540d9f602a01a524d776060d7c97eaede8b9e4d998f02310b49204f- Stable URL of this page
https://metraipolis.com/documento/dc_671d3f55cd3d
How to cite
TideRL: Boosting Agentic RL Goodput with Readiness-Aware Scheduling — arXiv:2608.10402 — arXiv (en), https://arxiv.org/abs/2608.10402 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_671d3f55cd3d (Metraipolis archive dc_671d3f55cd3d, sha256 6907619d6540d9f602a01a524d776060d7c97eaede8b9e4d998f02310b49204f)
BibTeX
@misc{dc671d3f55cd3d,
title = {TideRL: Boosting Agentic RL Goodput with Readiness-Aware Scheduling — arXiv:2608.10402},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.10402},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_671d3f55cd3d; sha256 6907619d6540d9f602a01a524d776060d7c97eaede8b9e4d998f02310b49204f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10402. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Reinforcement learning (RL) for large language models is moving toward multi-turn agentic workloads, where rollout tasks repeatedly pause for external environments, resume with growing contexts, and finish at highly variable times. In this setting, RL training goodput, measured by training throughput, matters more than raw GPU occupancy: GPU waiting and repeated prefill recomputation are pure overhead. We present TideRL, a readiness-aware elastic RL system with Continuous Task Batching, Resource-Aware Ref-Actor Pipelining, and Elastic Resource Scaling. CTB preserves useful rollout state, $\textrm{RA}^2\textrm{P}$ selects between decoupled streaming and colocated aggregation from the ready backlog and arrival interval, and ERS moves ranks between rollout and training using the same readiness signals. Across text-only and multi-modal agentic workloads, TideRL improves RL training goodput by up to 5.6$\times$ over synchronous baselines and over 33% over asynchronous baselines, while reaching similar task performance. It also improves KV cache hit rate by 1.58$\times$, reduces per-step training time by up to 44.3%, and cuts total waiting time by up to 77.6%.
Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-08-12 in cs.DC).
Categorie arXiv: Machine Learning (cs.LG) ; Distributed, Parallel, and Cluster Computing (cs.DC).
Pagina del lavoro: https://arxiv.org/abs/2608.10402 — 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 2026-10-19T06:09:08.099Z dalla pagina https://arxiv.org/catchup/cs.DC/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).