Analyzing the impact of CUDA versions on GPU applications

IF 2 4区计算机科学 Q2 COMPUTER SCIENCE, THEORY & METHODS

Parallel Computing Pub Date : 2024-02-29 DOI:10.1016/j.parco.2024.103081

Kohei Yoshida, Shinobu Miwa, Hayato Yamaki, Hiroki Honda

{"title":"Analyzing the impact of CUDA versions on GPU applications","authors":"Kohei Yoshida, Shinobu Miwa, Hayato Yamaki, Hiroki Honda","doi":"10.1016/j.parco.2024.103081","DOIUrl":null,"url":null,"abstract":"<div><p>CUDA toolkits are widely used to develop applications running on NVIDIA GPUs. They include compilers and are frequently updated to integrate state-of-the-art compilation techniques. Hence, many HPC users believe that the latest CUDA toolkit will improve application performance; however, considering results from CPU compilers, there are cases where this is not true. In this paper, we thoroughly evaluate the impact of CUDA toolkit version on the performance, power consumption, and energy consumption of GPU applications with four GPU architectures. Our results show that though the latest CUDA toolkit obtains the best performance, power consumption, and energy consumption for many applications in most cases, but we found a few exceptions. For such applications, we conducted an in-depth analysis using the SASS to identify why older CUDA toolkit achieve performance improvement. Our analysis showed that the factors that caused them are by three phenomena: aggressive loop unrolling, inefficient instruction scheduling, and the impact of host compilers.</p></div>","PeriodicalId":54642,"journal":{"name":"Parallel Computing","volume":"120 ","pages":"Article 103081"},"PeriodicalIF":2.0000,"publicationDate":"2024-02-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S016781912400019X/pdfft?md5=62bfbd6666b978a441b0d0daa8420592&pid=1-s2.0-S016781912400019X-main.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Parallel Computing","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S016781912400019X","RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, THEORY & METHODS","Score":null,"Total":0}

引用次数: 0

Abstract

CUDA toolkits are widely used to develop applications running on NVIDIA GPUs. They include compilers and are frequently updated to integrate state-of-the-art compilation techniques. Hence, many HPC users believe that the latest CUDA toolkit will improve application performance; however, considering results from CPU compilers, there are cases where this is not true. In this paper, we thoroughly evaluate the impact of CUDA toolkit version on the performance, power consumption, and energy consumption of GPU applications with four GPU architectures. Our results show that though the latest CUDA toolkit obtains the best performance, power consumption, and energy consumption for many applications in most cases, but we found a few exceptions. For such applications, we conducted an in-depth analysis using the SASS to identify why older CUDA toolkit achieve performance improvement. Our analysis showed that the factors that caused them are by three phenomena: aggressive loop unrolling, inefficient instruction scheduling, and the impact of host compilers.

查看原文本刊更多论文

分析 CUDA 版本对 GPU 应用程序的影响

CUDA 工具包被广泛用于开发在英伟达™（NVIDIA®）GPU 上运行的应用程序。这些工具包包括编译器，并经常更新，以整合最先进的编译技术。因此，许多高性能计算用户认为最新的CUDA工具包将提高应用程序的性能；然而，考虑到CPU编译器的结果，有些情况下并非如此。在本文中，我们全面评估了 CUDA 工具包版本对采用三种 GPU 架构的 GPU 应用程序的性能、功耗和能耗的影响。我们的结果表明，虽然最新的 CUDA 工具包在大多数情况下都能为许多应用获得最佳性能、功耗和能耗，但我们也发现了一些例外情况。针对这些应用，我们使用 SASS 进行了深入分析，以找出旧版 CUDA 工具包性能提升的原因。我们的分析表明，造成这些问题的因素有三个：激进的循环解卷、低效的指令调度以及主机编译器的影响。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Parallel Computing 工程技术-计算机：理论方法

CiteScore

3.50

自引率

7.10%

发文量

审稿时长

4.5 months

期刊介绍： Parallel Computing is an international journal presenting the practical use of parallel computer systems, including high performance architecture, system software, programming systems and tools, and applications. Within this context the journal covers all aspects of high-end parallel computing from single homogeneous or heterogenous computing nodes to large-scale multi-node systems. Parallel Computing features original research work and review articles as well as novel or illustrative accounts of application experience with (and techniques for) the use of parallel computers. We also welcome studies reproducing prior publications that either confirm or disprove prior published results. Particular technical areas of interest include, but are not limited to: -System software for parallel computer systems including programming languages (new languages as well as compilation techniques), operating systems (including middleware), and resource management (scheduling and load-balancing). -Enabling software including debuggers, performance tools, and system and numeric libraries. -General hardware (architecture) concepts, new technologies enabling the realization of such new concepts, and details of commercially available systems -Software engineering and productivity as it relates to parallel computing -Applications (including scientific computing, deep learning, machine learning) or tool case studies demonstrating novel ways to achieve parallelism -Performance measurement results on state-of-the-art systems -Approaches to effectively utilize large-scale parallel computing including new algorithms or algorithm analysis with demonstrated relevance to real applications using existing or next generation parallel computer architectures. -Parallel I/O systems both hardware and software -Networking technology for support of high-speed computing demonstrating the impact of high-speed computation on parallel applications