GEP在磁盘负载均衡方面的应用研究

如何提高存储子系统的 I/O性能一直以来都是计算机领域的一个研究热点 ,而目前提高存储子系统的 I/O性能的一个最大障碍就是负载不均衡。提出了一种采用基因表达式编程 ( GEP)来实现基于分条技术的磁盘动态负载均衡的算法。该方法包括基于分条技术的文件划分算法和为实现负载均衡的文件分配算法。该算法采用多基因家族结构的染色体编码来表示物理磁盘组与逻辑磁盘的映射关系。在操作上 ,采用选择复制、倒置和交换等特殊的搜索算子。

Research of disk load balancing based on GEP

With the increase of the number of disks in storage subsystems due to rapidly increasing capacity requirements, the largest performance problem of the storage is load imbalance. This paper presented a new scheme based on disk striping and GEP to solve the problem, including the file partition algorithm based on disk striping and the file allocation algorithm for load balance. This algorithm represented the chromosome of the tree with the multigene family structure coding. And applied three combination-specific genetic operators: selection, inversion and permutation.