File System Write-Optimization on High-Performance Computing Support Platform

DOI

10.2991/icmse-17.2017.29How to use a DOI?

Keywords

Write Optimization; Distributed File System; Write-Ahead Log; HCSP

Abstract

The development of energy grid results in explosive data increasing. Handling the large volume of data not only requires vast storage space, but also requires the power to process them. Based on various write-optimization technologies, we proposed the high-performance computing support platform (HCSP) as a demonstration of write-optimization performance. HCSP integrates both hot/cold data layering and full-stack sense optimization, utilizes both SSD and SATA disk to accelerate the data storage and access. The experimental results show that compare to the small servers, HCSP not only provides a high-performance data access but also have a relatively short response time in various typical scenarios.

Copyright

© 2017, 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  - Jiye Wang
AU  - Nan Zeng
AU  - Jun Yu
PY  - 2017/04
DA  - 2017/04
TI  - File System Write-Optimization on High-Performance Computing Support Platform
BT  - Proceedings of the 2017 7th International Conference on Manufacturing Science and Engineering (ICMSE 2017)
PB  - Atlantis Press
SP  - 156
EP  - 159
SN  - 2352-5401
UR  - https://doi.org/10.2991/icmse-17.2017.29
DO  - 10.2991/icmse-17.2017.29
ID  - Wang2017/04
ER  -