Microarchitectural performance comparison of Intel Knights Corner and Intel Sandy Bridge with CFD applications

被引:8
作者
Che, Yonggang [1 ]
Zhang, Lilun [1 ]
Wang, Yongxian [1 ]
Xu, Chuanfu [1 ]
Liu, Wei [1 ]
Wang, Zhenghua [1 ]
机构
[1] Natl Univ Def Technol, Sci & Technol Parallel & Distributed Proc Lab, Changsha, Hunan, Peoples R China
基金
中国国家自然科学基金;
关键词
Many Integrated Core; Multicore processors; CFD applications; Performance comparison; Microarchitecture factors; Application characteristics; COMPACT NONLINEAR SCHEMES; HIGH-ORDER; INTERFACE;
D O I
10.1007/s11227-014-1245-3
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
This paper comparatively evaluates the microarchitectural performance of two representative Computational Fluid Dynamics (CFD) applications on the Intel Many Integrated Core (MIC) product, the Intel Knights Corner (KNC) coprocessor, and the Intel Sand Bridge (SNB) processor. Performance Monitoring Unit-based measurement method is used, along with a two-phase measurement method and some considerations to minimize the errors and instabilities. The results show that the CFD applications are sensitive to architecture factors. Their single thread performance and efficiency on KNC are much lower than that on SNB. Branch prediction and memory access are two primary factors that make the performance difference. The applications' low-computational intensity and inefficient vector instruction usage are two additional factors. To be more efficient for the CFD applications, the MIC architecture needs to improve its branch prediction mechanism and memory hierarchy. Fine tuning of application codes is also crucial and is hard work.
引用
收藏
页码:321 / 348
页数:28
相关论文
共 30 条
[1]  
[Anonymous], P TACC INT HIGHL PAR
[2]  
[Anonymous], INT 64 IA 32 ARCH SO
[3]  
[Anonymous], ACM SIGMETRICS PERFO
[4]   A portable programming interface for performance evaluation on modern processors [J].
Browne, S ;
Dongarra, J ;
Garner, N ;
Ho, G ;
Mucci, P .
INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS, 2000, 14 (03) :189-204
[5]  
Cadambi S, 2013, HPDC 13 NEW YORK NY, P215
[6]  
Che Y-G, 2012, ANN C HIGH PERF COMP, P39
[7]  
Cramer T, 2012, MAN COR APPL RES COM, P38
[8]   Developing high-order weighted compact nonlinear schemes [J].
Deng, XG ;
Zhang, HX .
JOURNAL OF COMPUTATIONAL PHYSICS, 2000, 165 (01) :22-44
[9]   High-Order and High Accurate CFD Methods and Their Applications for Complex Grid Problems [J].
Deng, Xiaogang ;
Mao, Meiliang ;
Tu, Guohua ;
Zhang, Hanxin ;
Zhang, Yifeng .
COMMUNICATIONS IN COMPUTATIONAL PHYSICS, 2012, 11 (04) :1081-1102
[10]   Extending Weighted Compact Nonlinear Schemes to Complex Grids with Characteristic-Based Interface Conditions [J].
Deng, Xiaogang ;
Mao, Meiliang ;
Tu, Guohua ;
Zhang, Yifeng ;
Zhang, Hanxin .
AIAA JOURNAL, 2010, 48 (12) :2840-2851