首页 | 官方网站   微博 | 高级检索  
     

NAS设备的磁盘调度模型分析
引用本文:胡鸣,周敬利,余胜生.NAS设备的磁盘调度模型分析[J].小型微型计算机系统,2004,25(9):1578-1581.
作者姓名:胡鸣  周敬利  余胜生
作者单位:华中科技大学,计算机学院,国家外存储实验室,湖北,武汉,430074
基金项目:国家自然科学基金 ( 60 3 73 0 88)资助,国防科技预研基金项目 ( 4 13 160 5 0 2 )的资助
摘    要:附网存储 (NAS)设备的性能目标是优化网络存储数据访问和存储子系统的管理 .本文旨在显示随着磁盘转速的增加 ,NAS应该从磁盘硬件的最佳工作性能出发 ,整体配合以提高它的 I/O操作性能 .为了发掘 NAS最佳的工作性能 ,我们建立同时基于磁臂位置和旋转位置的精确的磁盘调度模型 ,并提出测量调度算法性能的方法 .以 HP975 6 0性能参数为基础 ,进行理论分析和模拟测试 .结果磁盘转速越快 ,磁盘访问的开销越大 .因此 ,NAS设备的设计必须从整体上考虑磁盘调度策略的选择 ,并行多磁盘结构的选择 ,文件 cache的分配和文件系统布局等 ,以便提高网络存储数据访问的性能

关 键 词:附网存储(NAS)  基于磁臂位置和旋转位置的磁盘调度  磁盘调度模型
文章编号:1000-1220(2004)09-1578-04

Analysis of Disk Scheduling Model Based on NAS Devices
HU Ming,ZHOU Jing li,YU Sheng sheng.Analysis of Disk Scheduling Model Based on NAS Devices[J].Mini-micro Systems,2004,25(9):1578-1581.
Authors:HU Ming  ZHOU Jing li  YU Sheng sheng
Abstract:The perfomance goal of a network attached storage (NAS) device is to optimize netwrok storage data access and the management of its storage subsystem. This paper attempts to show that, as disk rotational speed increases, NAS should base on the optimal working performance in disk hardwre and make all its components cooperate in order to impove the performance of its I/O operations. To develop the optimal working performance in NAS,we establish the exact disk scheduling model based on both disk arm and rotational position, and put forward the method about how to measure the performance of the disk scheduling algorithms. Based on the performance parameters of HP97560, we make the theoretic analysis and the simulation test. As a result, the faster the rotational speed, the higher the overhead of disk access. Therefore, the design of a NAS device must wholly consider the choice of disk scheduling policies and of parallel multiple disk structure, the distribution of file cache, file system layout, and so on in order to improve the performance of network storage data access.
Keywords:network  attached storage(NAS)  disk  scheduling basded on disk arm and rotational position  disk  scheduling model
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号