Metraipolis › archive › dc_90f3aea09eda
GAUSS: A Workload-Aware Analytical and Monte-Carlo Simulator for LLM Inference Serving — arXiv:2610.37989
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.37989
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T06:08:42.583Z
- SHA-256 of the archived text
5a148d567771a46b625a6e07bfe6cc0b637224a3ad6406f242b17f2691109e10- Stable URL of this page
https://metraipolis.com/documento/dc_90f3aea09eda
How to cite
GAUSS: A Workload-Aware Analytical and Monte-Carlo Simulator for LLM Inference Serving — arXiv:2610.37989 — arXiv (en), https://arxiv.org/abs/2610.37989 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_90f3aea09eda (Metraipolis archive dc_90f3aea09eda, sha256 5a148d567771a46b625a6e07bfe6cc0b637224a3ad6406f242b17f2691109e10)
BibTeX
@misc{dc90f3aea09eda,
title = {GAUSS: A Workload-Aware Analytical and Monte-Carlo Simulator for LLM Inference Serving — arXiv:2610.37989},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.37989},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_90f3aea09eda; sha256 5a148d567771a46b625a6e07bfe6cc0b637224a3ad6406f242b17f2691109e10},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.37989. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language model (LLM) inference has become an operational workload that fleets must provision, monitor, and bill against latency service-level objectives (SLOs). A single throughput number cannot answer the questions operators actually face, because generative serving couples two phases with very different service characteristics, continuous batching reshapes the effective service process, and user-visible tail latency can degrade long before aggregate throughput saturates. We present GAUSS, a workload-aware analytical and Monte-Carlo simulator that predicts the full distributions of time-to-first-token (TTFT) and time-between-tokens (TBT), together with the sustainable request rate, for a given workload, engine, and accelerator. GAUSS pairs calibrated, phase-specific latency surfaces for prefill and decode with a Lindley-recurrence model of queueing under stochastic arrivals and estimates performance by ensemble Monte-Carlo sampling rather than by replaying a single trace. This lets GAUSS reason about workloads, routing policies, and hardware that have never been deployed, and to capture the latency tails that trace-driven simulators systematically under-sample. We describe the model and show how it supports operating-point selection, traffic segmentation, and shortest-remaining-processing-time scheduling capacity questions that trace-driven simulators cannot address. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.37989 — 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 2026-10-31T06:07:08.445Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).