arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.03583cs.DC
收藏

Loki:一个支持硬件与精度扩展的机器学习推理流水线服务系统

Loki: A System for Serving ML Inference Pipelines with Hardware and Accuracy Scaling

Sohaib Ahmad, Hui Guan, Ramesh K. Sitaraman

AI总结:

本文提出推理流水线服务系统Loki,通过结合硬件扩展与精度扩展、最优资源分配理论框架及查询路由算法,在满足吞吐量需求的同时显著提升系统容量并大幅降低SLO违规。

摘要:

机器学习(ML)的快速普及凸显了以高吞吐量和高资源效率服务ML模型的重要性。传统上,应对不断增长的查询需求的方法主要集中在硬件扩展上,即增加服务器数量或计算能力。然而,由于可用预算或计算资源的限制,这种策略往往不切实际。作为一种替代方案,精度扩展通过调整ML模型的精度来适应波动的查询需求,提供了一种有前景的解决方案。然而,现有的精度扩展技术针对的是独立的ML模型,在管理推理流水线时往往表现不佳。此外,它们缺乏与硬件扩展的集成,导致低需求期间可能出现资源效率低下的问题。为了解决这些局限性,本文介绍了Loki,一个旨在通过硬件和精度扩展有效服务推理流水线的系统。Loki融合了一个用于最优资源分配的创新理论框架和一个有效的查询路由算法,旨在提高系统精度并最小化延迟截止时间违规。我们的实证评估表明,通过精度扩展,固定规模集群的有效容量相比仅依赖硬件扩展可以提高超过$2.7\ imes$。与最先进的推理服务系统相比,Loki在服务级别目标(SLO)违规方面实现了高达$10\ imes$的减少,同时以最小的精度妥协满足了吞吐量需求。

英文摘要:

The rapid adoption of machine learning (ML) has underscored the importance of serving ML models with high throughput and resource efficiency. Traditional approaches to managing increasing query demands have predominantly focused on hardware scaling, which involves increasing server count or computing power. However, this strategy can often be impractical due to limitations in the available budget or compute resources. As an alternative, accuracy scaling offers a promising solution by adjusting the accuracy of ML models to accommodate fluctuating query demands. Yet, existing accuracy scaling techniques target independent ML models and tend to underperform while managing inference pipelines. Furthermore, they lack integration with hardware scaling, leading to potential resource inefficiencies during low-demand periods. To address the limitations, this paper introduces Loki, a system designed for serving inference pipelines effectively with both hardware and accuracy scaling. Loki incorporates an innovative theoretical framework for optimal resource allocation and an effective query routing algorithm, aimed at improving system accuracy and minimizing latency deadline violations. Our empirical evaluation demonstrates that through accuracy scaling, the effective capacity of a fixed-size cluster can be enhanced by more than $2.7\times$ compared to relying solely on hardware scaling. When compared with state-of-the-art inference-serving systems, Loki achieves up to a $10\times$ reduction in Service Level Objective (SLO) violations, with minimal compromises on accuracy and while fulfilling throughput demands.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑