arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.10837cs.DCcs.AIcs.ARcs.PF
收藏

BayesPerf:使用贝叶斯统计最小化性能监视错误

BayesPerf: Minimizing Performance Monitoring Errors Using Bayesian Statistics

University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Subho S. Banerjee, Saurabh Jha, Zbigniew T. Kalbarczyk, Ravishankar K. Iyer

摘要:

硬件性能计数器(HPC)测量低级架构和微架构事件,提供有关系统状态的动态上下文信息。然而,HPC测量由于非确定性(例如,由于事件多路复用或OS中断处理行为而导致的计数不足)。在本文中,我们提出了BayesPerf,一个系统,用于量化HPC测量的不确定性,通过使用域驱动的贝叶斯模型,捕获HPC之间的微架构关系,共同推断其值的概率分布。我们提供了一个加速器的设计和实现,该加速器允许x86和ppc64 CPU的BayesPerf模型的低延迟和低功耗推理。当事件被多路复用时,BayesPerf将HPC测量的平均误差从40.1%降低到7.6%。BayesPerf在实时决策中的价值通过PCIe传输调度的简单示例来说明。

英文摘要:

Hardware performance counters (HPCs) that measure low-level architectural and microarchitectural events provide dynamic contextual information about the state of the system. However, HPC measurements are error-prone due to non determinism (e.g., undercounting due to event multiplexing, or OS interrupt-handling behaviors). In this paper, we present BayesPerf, a system for quantifying uncertainty in HPC measurements by using a domain-driven Bayesian model that captures microarchitectural relationships between HPCs to jointly infer their values as probability distributions. We provide the design and implementation of an accelerator that allows for low-latency and low-power inference of the BayesPerf model for x86 and ppc64 CPUs. BayesPerf reduces the average error in HPC measurements from 40.1% to 7.6% when events are being multiplexed. The value of BayesPerf in real-time decision-making is illustrated with a simple example of scheduling of PCIe transfers.

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。

↑