Solving Parity Declustering Optimal Problems in Network Based RAID by Using Two-Weighted-Objective Genetic Algorithms
Yang Mi · 2005
RAID,a technology to improve the reliability and performance of storage system,has been got a widely us- ing. Although the researchers about the structure of RAID and data layout are very popular, the researchers about the data layout of network based RAID are little. In this paper, first, we mainly introduce some knowledge about parity declustering and genetic algorithms, present a new thought to solving parity declustering optimal problems in the network based RAID by two-weighted-objective genetic algorithms. Then we take Balance reconstruction work- load and Balanced parity overheadas two objectives, used the changed NSGA to solving parity declustering optimal problems in the network based RAID system. At last we describe the experimental results.