Extreme-scale parallel computing: bottlenecks and strategies
| Autor: | Ze-yao Mo |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Computer Networks and Communications
Computer science Scale (chemistry) Synchronizing 010103 numerical & computational mathematics 02 engineering and technology Parallel computing Supercomputer Key issues 01 natural sciences Programming productivity Hardware and Architecture Extreme scale Signal Processing 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Numerical computing 0101 mathematics Electrical and Electronic Engineering |
| Zdroj: | Frontiers of Information Technology & Electronic Engineering. 19:1251-1260 |
| ISSN: | 2095-9230 2095-9184 |
| DOI: | 10.1631/fitee.1800421 |
| Popis: | 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. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink