An optimal candidate selection model for self-acting load balancing of parallel file system

Authors:
Bin Dong;Xiuqiao Li;Limin Xiao;Li Ruan
Affiliations:
School of Computer Science and Engineering, Beihang University, Beijing 100191, China.;School of Computer Science and Engineering, Beihang University, Beijing 100191, China.;School of Computer Science and Engineering, Beihang University, Beijing 100191, China.;School of Computer Science and Engineering, Beihang University, Beijing 100191, China
Venue:
International Journal of High Performance Computing and Networking
Year:
2012

Citing 6
Cited 0

The art of computer programming, volume 3: (2nd ed.) sorting and searching

The art of computer programming, volume 3: (2nd ed.) sorting and searching
File Assignment in Parallel I/O Systems with Minimal Variance of Service Time

IEEE Transactions on Computers
Data partitioning and load balancing in parallel disk systems

The VLDB Journal — The International Journal on Very Large Data Bases
PVFS: a parallel file system for linux clusters

ALS'00 Proceedings of the 4th annual Linux Showcase & Conference - Volume 4
Characterizing and predicting the I/O performance of HPC applications using a parameterized synthetic benchmark

Proceedings of the 2008 ACM/IEEE conference on Supercomputing
Self-Acting Load Balancing with Parallel Sub File Migration for Parallel File System

CSO '10 Proceedings of the 2010 Third International Joint Conference on Computational Science and Optimization - Volume 02

Quantified Score

Hi-index	0.00

Visualization

Abstract

The load imbalance among the I/O servers of parallel file systems is an emerging challenge to effectively utilise the high-end computing system. However, most contemporary parallel file systems lack the ability to handle the load imbalance among the I/O servers. In order to address this problem, a self-acting load balancing strategy is proposed in this paper. The self-acting load balancing is based on two novel components: a statistic model which measures the relationship between the response time and the load of I/O server and an optimal candidate selection model for the dynamic file migration. The experimental results show that the parallel file system with the proposed load balancing scheme can improve the performance of parallel I/O.