The Long and Winding Road Toward Efficient High-Performance Computing

被引:1
作者
Jalby, William [1 ]
Kuck, David [2 ]
Malony, Allen D. [3 ]
Masella, Michel [4 ]
Mazouz, Abdelhafid [5 ]
Popov, Mihail [1 ,6 ]
机构
[1] Univ Versailles St Quentin En Yvelines, Dept Comp Sci, UFR Sci, F-78035 Versailles, France
[2] INTEL, SSG, Austin, TX 78746 USA
[3] Univ Oregon, Dept Comp Sci, Eugene, OR 97403 USA
[4] CEA, DRF DSV, F-91191 Gif Sur Yvette, France
[5] HPC, BULL ATOS, F-91680 Bruyeres Le Chatel, France
[6] Uppsala Univ, Dept Informat Technol, S-75105 Uppsala, Sweden
关键词
Autotuning; benchmarking; hardware design; molecular dynamics; performance evaluation tools; MOLECULAR-DYNAMICS; IMPLEMENTATION; CODE; DESIGN;
D O I
10.1109/JPROC.2018.2851190
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The major challenge to Exaflop computing, and more generally, efficient high-end computing, is in finding the best "matches" between advanced hardware capabilities and the software used to program applications, so that top performance will be achieved. Several benchmarks show very disappointing performance progress over the last decade, clearly indicating a mismatch between hardware and software. To remedy this problem, it is important that key performance enablers at the software level-autotuning, performance analysis tools, full application optimization-are understood. For each area, we highlight major limitations and most promising approaches to reaching better performance and energy levels. Finally, we conclude by analyzing hardware and software design, trying to pave the way for more tightly integrated hardware and software codesign.
引用
收藏
页码:1985 / 2003
页数:19
相关论文
共 61 条
[1]   HPCTOOLKIT: tools for performance analysis of optimized parallel programs [J].
Adhianto, L. ;
Banerjee, S. ;
Fagan, M. ;
Krentel, M. ;
Marin, G. ;
Mellor-Crummey, J. ;
Tallent, N. R. .
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2010, 22 (06) :685-701
[2]  
Agakov F, 2006, INT SYM CODE GENER, P295
[3]  
[Anonymous], 1998, SC 98, DOI [10.5555/509058.509096, DOI 10.1109/SC.1998.10004]
[4]  
[Anonymous], 2011, P 2011 INT C HIGH PE
[5]  
[Anonymous], 1999, P DEP DEF HPCMP US G
[6]  
[Anonymous], 2005, ACMIEEE SC 2005 C SC
[7]  
[Anonymous], 2014, 2014 21 INT C HIGH P, DOI DOI 10.1109/HIPC.2014.7116904
[8]  
[Anonymous], P 2010 C PAR PROC EU
[9]  
[Anonymous], 2000, SC 00, DOI [DOI 10.1109/SC.2000.10052, 10.1109/SC.2000.10052]
[10]  
BENDIFALLAH Z., 2014, Tools for High Performance Computing 2013, P107