Research on Computer Architecture based on Distributed Cluster Structure

DOI

10.2991/icecee-15.2015.258How to use a DOI?

Keywords

computer architecture; distributed cluster structure; high performance computing (HPC); CPU and GPU.

Abstract

With the fast development of image science and embedded system technique, it is widely accepted that future HPC systems will be limited by their power consumption. The current high performance computing system is a commodity server processors, design for many years to achieve maximum performance, and then it dawned on energy efficiency. In this paper we advocate a different approach: computer architecture based on distributed cluster structure. We introduce the architecture of Tibidabo, the first large-scale HPC cluster built from ARM multicore chips, and a detailed performance and energy efficiency evaluation. We now design experience and to improve the energy efficiency of future HPC systems based on the low power core. The experimental result shows the effectiveness of our methodology, further in-depth research portion is also discussed with case studies.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yuwen Zheng
PY  - 2015/06
DA  - 2015/06
TI  - Research on Computer Architecture based on Distributed Cluster Structure
BT  - Proceedings of the 2015 International Conference on Electrical, Computer Engineering and Electronics
PB  - Atlantis Press
SP  - 1380
EP  - 1385
SN  - 2352-538X
UR  - https://doi.org/10.2991/icecee-15.2015.258
DO  - 10.2991/icecee-15.2015.258
ID  - Zheng2015/06
ER  -