arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2410.22229cs.NIcs.CL
收藏

利用性能预测加速SoC智能网卡上的有状态网络应用

Accelerating Stateful Network Applications with Performance Prediction on SoC SmartNICs

Zhejiang University(浙江大学);Alibaba Cloud(阿里云);Northeastern University(东北大学);Harvard University(哈佛大学)

Shaoke Xi, Jiaqi Gao, Fuliang Li, Minlan Yu, Ennan Zhai

AI总结:

Vela框架通过预测性编译器和轻量级运行时,解决SoC智能网卡上有状态网络功能卸载的资源分配与动态过载问题,显著降低CPU使用率并提升吞吐量。

摘要:

将有状态网络功能卸载到多线程SoC智能网卡上,有望带来显著的性能和成本优势。然而,实现这一潜力受到两个基本挑战的阻碍。首先,在没有性能指导的情况下,开发者被迫陷入缓慢的手动试错循环,通过部署和测试来寻找可行的资源分配方案。其次,在流量变化的情况下维持性能,需要适应状态驻留并在编译时固定的内存布局内处理过载。本文介绍了Vela,一个通过模型驱动的编译时/运行时协同设计来应对这两个挑战的框架。其核心是一个预测性编译器,用快速、自动化的分析取代手动调优循环,使用一种新颖的、以状态为中心的分析模型来估计任何给定资源分配方案的吞吐量上限。这辅以一个轻量级运行时,在编译后的内存布局内动态管理缓存内容并缓解过载。我们在Netronome Agilio和NVIDIA BlueField 3智能网卡上,跨四个网络功能实现了Vela并进行了评估。相对于实现相同吞吐量的最佳基线,Vela将主机CPU或板载Arm核心使用率降低了62.9%-91.9%。对于NATLB工作负载,在相同的主机/Arm核心数量下,Vela的吞吐量也比最佳基线提高了15.2%-120.8%。

英文摘要:

Offloading stateful network functions to multi-threaded SoC SmartNICs promises significant performance and cost benefits. However, realizing this potential is hindered by two fundamental challenges. First, without performance guidance, developers are forced into a slow, manual trial-and-error cycle of deploying and testing to find a feasible resource allocation. Second, sustaining performance under changing traffic requires adapting state residency and handling overload within memory layouts fixed at compile time. This paper introduces Vela, a framework that addresses both challenges through a model-driven, compile-time/runtime co-design. Its core is a predictive compiler that replaces the manual tuning loop with fast, automated analysis, using a novel, state-centric analytical model to estimate the throughput ceiling of any given resource allocation plan. This is complemented by a lightweight runtime that dynamically manages cache contents within the compiled memory layout and mitigates overload. We implement and evaluate Vela on Netronome Agilio and NVIDIA BlueField 3 SmartNICs across four NFs. Vela reduces host CPU or on-board Arm core usage by 62.9%-91.9% relative to the best baseline achieving the same throughput. For the NATLB workload, vela also improves throughput by 15.2%-120.8% over the best baseline at the same host/Arm core count.

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。

↑