Extreme-scale parallel computing:bottlenecks and strategies  

超大规模并行计算:瓶颈与对策（英文）

作　　者：Ze-yao MO 

机构地区：[1]CAEP Software Center for High Performance Numerical Simulation [2]Institute of Applied Physics and Computational Mathematics

出　　处：《Frontiers of Information Technology & Electronic Engineering》2018年第10期1251-1260,共10页信息与电子工程前沿（英文版）

基　　金：Project supported by the National Natural Science Foundation of China(No.91430218);the National Key Technology R&D Program of China(Nos.2016YFB0201300 and 2017YFB0202103)

摘　　要：Extreme-scale numerical simulations seriously demand extreme parallel computing capabilities. To address the challenges of these capabilities toward exascale, we systematically analyze the major bottlenecks of parallel computing research from three perspectives： computational scale, computing efficiency, and programming productivity. For these bottlenecks, we propose a series of urgent key issues and coping strategies. This study will be useful in synchronizing development between the numerical computing capability and supercomputer peak performance.超大规模数值模拟极大依赖并行计算能力。从计算规模、计算效率和编程生产率3个维度,系统分析了超大规模并行计算能力的主要瓶颈,提出亟待研究的若干关键技术问题和技术对策。本文对推动数值模拟软件计算能力与超级计算机峰值性能的同步提升具有参考价值。

关 键 词：EXTREME SCALE Numerical simulation Parallel COMPUTING SUPERCOMPUTERS 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]