Mining Blocks' Association Rules for Disk Data Perfecting
Authors
L.Y Zhu, G.Q Xiao, J.W Liao
Corresponding Author
L.Y Zhu
Available Online June 2015.
- DOI
- 10.2991/cisia-15.2015.113How to use a DOI?
- Keywords
- association rule; prefetching; storage system
- Abstract
As the processor-I/O gap continues widening, the modern I/O-bound applications such as Online Transaction Processing (OLTP) often suffer from great latency of disk access. In this paper, we argue that the block association rule can be recognized as common semantic information embedded in disk I/O traces and it can be utilized to direct disk prefetching. We propose a novel mining approach to extract such association rules to benefit prefetching. The experimental results show that this proposed approach is effective in increasing prefetching accuracy, and the disk latency can be reduced by this association rule based prefetching.
- 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/).
Cite this article
TY - CONF AU - L.Y Zhu AU - G.Q Xiao AU - J.W Liao PY - 2015/06 DA - 2015/06 TI - Mining Blocks' Association Rules for Disk Data Perfecting BT - Proceedings of the International Conference on Computer Information Systems and Industrial Applications PB - Atlantis Press SP - 418 EP - 421 SN - 2352-538X UR - https://doi.org/10.2991/cisia-15.2015.113 DO - 10.2991/cisia-15.2015.113 ID - Zhu2015/06 ER -