arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.02368cs.DCcs.ARcs.GR
收藏

并行GPU内核功耗-性能特性的经验统计方法

An Empirical-cum-Statistical Approach to Power-Performance Characterization of Concurrent GPU Kernels

Nilanjan Goswami, Amer Qouneh, Chao Li, Tao Li

摘要:

数据中心中功率和能量高效的吞吐量加速器(GPU)的日益增长的部署要求增强GPU的功率-性能协同优化能力。使用加速器实现百亿亿次计算需要进一步提高功率效率。通过加速器中的硬连线内核并发启用,工作负载间和工作负载内的同步内核计算预测以较低的能量预算增加吞吐量。为了提高体系结构的每瓦性能指标,需要对真实世界的吞吐量工作负载(具有并发内核执行)进行系统的实证研究。为此,我们提出了一个多内核的吞吐量工作负载生成框架,这将有利于积极的能源和性能管理的exascale数据中心,并将刺激协同功率性能协同优化的吞吐量架构。此外,我们展示了一个多内核的吞吐量基准测试套件的基础上的框架,封装对称,非对称和共存(经常出现在一起)的内核为基础的工作负载。平均而言,我们的分析表明,吞吐量架构中内核执行的空间和时间并发在GTX 470,Tesla M2050和Tesla K20中的12个基准测试中分别节省了32%,26%和33%的能耗。并发性和提高的利用率通常是相关的,但并不意味着功耗有显著的差异。多样性分析建议的多内核确认特性的变化和功率分布的多样性套件内。此外,我们解释了几个发现的并发吞吐量工作负载的功率性能协同优化。

英文摘要:

Growing deployment of power and energy efficient throughput accelerators (GPU) in data centers demands enhancement of power-performance co-optimization capabilities of GPUs. Realization of exascale computing using accelerators requires further improvements in power efficiency. With hardwired kernel concurrency enablement in accelerators, inter- and intra-workload simultaneous kernels computation predicts increased throughput at lower energy budget. To improve Performance-per-Watt metric of the architectures, a systematic empirical study of real-world throughput workloads (with concurrent kernel execution) is required. To this end, we propose a multi-kernel throughput workload generation framework that will facilitate aggressive energy and performance management of exascale data centers and will stimulate synergistic power-performance co-optimization of throughput architectures. Also, we demonstrate a multi-kernel throughput benchmark suite based on the framework that encapsulates symmetric, asymmetric and co-existing (often appears together) kernel based workloads. On average, our analysis reveals that spatial and temporal concurrency within kernel execution in throughput architectures saves energy consumption by 32%, 26% and 33% in GTX470, Tesla M2050 and Tesla K20 across 12 benchmarks. Concurrency and enhanced utilization are often correlated but do not imply significant deviation in power dissipation. Diversity analysis of proposed multi-kernels confirms characteristic variation and power-profile diversity within the suite. Besides, we explain several findings regarding power-performance co-optimization of concurrent throughput workloads.

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。

↑