面向CFD应用的Intel持久内存性能评估

计算机工程与科学 ›› 2022, Vol. 44 ›› Issue (09): 1550-1556.

面向CFD应用的Intel持久内存性能评估

文敏华1,陈江2,胡广超1,韦建文1,王一超1,林新华1

(1.上海交通大学网络信息中心，上海 200240;2.英特尔（中国）有限公司，北京 100013)

收稿日期:2022-01-15 修回日期:2022-05-09 接受日期:2022-09-25 出版日期:2022-09-25 发布日期:2022-09-25
基金资助:
国家重点研发计划（2018YFA0404603）；上海交通大学“转化医学交叉研究基金”（ZH2018ZDA02）

Performance evaluation of Intel persistent memory for CFD applications

WEN Min-hua1,CHEN Jiang2,HU Guang-chao1,WEI Jian-wen1,WANG Yi-chao1,LIN Xin-hua1

(1.Network & Information Center，Shanghai Jiao Tong University，Shanghai 200240;
2.Intel China Ltd.，Beijing 100013，China)

Received:2022-01-15 Revised:2022-05-09 Accepted:2022-09-25 Online:2022-09-25 Published:2022-09-25

摘要/Abstract

摘要： 在科学计算领域，数据规模随着数值模拟精度要求的提高而快速增长，以DRAM为主存的传统方案由于成本高而难以扩展容量，近年来越来越被关注的持久内存技术有望解决这一问题。持久内存是在DRAM和SSD之间的补充，相比DRAM，持久内存具有容量大、性价比高的优点，但是性能也相对较低。为测试持久内存的应用性能，面向科学计算的一个重要领域——计算流体力学（CFD），对Intel持久内存进行性能评估。实验中，持久内存采用了最易于使用的内存模式，源码不需要任何修改，测试程序涵盖内存基准测试和3种常见的CFD算法，实验结果表明，在内存模式下，对不同CFD算法，相比纯DRAM的配置，持久内存的引入会带来一定的性能损失，且该损失随数据规模的增加而增大；另一方面，持久内存的部署使单服务器能支撑超大数据规模的数值模拟。

关键词: 计算流体力学, 持久内存, 性能评估

Abstract: In the field of scientific computing, the amount of data is increasing rapidly with the increase in the accuracy of numerical simulations. The traditional memory solution based on DRAM is difficult to expand the capacity due to the high cost. In recent years, the persistent memory technology has attracted more and more attentions and is expected to solve this problem. Persistent memory is a supplement between DRAM and SSD. Compared with DRAM, persistent memory has the advantages of large capacity and high cost performance, but its performance is relatively lower. To test the application performance of persistent memory, we evaluate the performance of Intel persistent memory for Computational Fluid Dynamics (CFD), an important area of scientific computing. In the experiment, persistent memory adopts the most easy-to-use memory mode, the source code does not need any modification, and the test program covers memory benchmark test and 3 common CFD algorithms. The experimental results show that in the memory mode, for different CFD algorithms, compared with DRAM as main memory, the introduction of persistent memory brings some performance loss, which increases with the increase of the data size. On the other hand, the deployment of persistent memory enables a single server to support numerical simulations with extremely large size of data.

Key words: computational fluid dynamics (CFD), persistent memory, performance evaluation ,

文敏华, 陈江, 胡广超, 韦建文, 王一超, 林新华. 面向CFD应用的Intel持久内存性能评估[J]. 计算机工程与科学, 2022, 44(09): 1550-1556.

WEN Min-hua, CHEN Jiang, HU Guang-chao, WEI Jian-wen, WANG Yi-chao, LIN Xin-hua. Performance evaluation of Intel persistent memory for CFD applications[J]. Computer Engineering & Science, 2022, 44(09): 1550-1556.

参考文献［13］

［1］	Anderson J D,Wendt J.Computational fluid dynamics［M］.New York:McGraw-Hill,1995.
［2］	Piomelli U.Large-Eddy simulation:Achievements and challenges［J］.Progress in Aerospace Sciences,1999,35(4):335-362.
［3］	Moin P,Mahesh K.Direct numerical simulation:A tool in turbulence research［J］.Annual Review of Fluid Mechanics,1998,30(1):539-578.
［4］	Weiland M,Brunst H,Quintino T,et al.An early evaluation of Intel’s Optane DC persistent memory module and its impact on high-performance scientific applications［C］∥Proc of the International Conference for High Performance Comput- ing,Networking,Storage and Analysis,2019:1-19.
［5］	Yang J,Kim J,Hoseinzadeh M,et al.An empirical guide to the behavior and use of scalable persistent memory［C］∥Proc of the 18th Conference on File and Storage Technologies,2020:169-182.
［6］	Akram S.Performance evaluation of Intel optane memory for managed workloads［J］.ACM Transactions on Architecture and Code Optimization,2021,18(3):1-26.
［7］	Mason T,Doudali T D,Seltzer M,et al.Unexpected performance of Intel OptaneTM DC persistent memory［J］.IEEE Computer Architecture Letters,2020,19(1):55-58.
［8］	Patil O,Ionkov L,Lee J,et al.Performance characterization of a DRAM-NVM hybrid memory architecture for HPC applications using Intel Optane DC persistent memory modules［C］∥Proc of the International Symposium on Memory Systems,2019:288-303.
［9］	Hady F T,Foong A,Veal B,et al.Platform storage performance with 3D XPoint technology［J］.Proceedings of the IEEE,2017,105(9):1822-1833.
［10］	https:∥www.cs.virginia.edu/stream/.
［11］	Mallinson A C, Beckingsale D A, Gaudin W P, et al. Cloverleaf:Preparing hydrodynamics codes for exascale［C］∥Proc of Cray User Group 2013,2013:1.
［12］	McIntosh-Smith S,Martineau M,Deakin T,et al.TeaLeaf:A mini-application to enable design-space explorations for iterative sparse linear solvers［C］∥Proc of 2017 IEEE International Conference on Cluster Computing,2017:842-849.
［13］	Bird G A.Molecular gas dynamics［M］.Oxford:Clarendon Press,1976.