首页 | 本学科首页   官方微博 | 高级检索  
     

基于智能存储设备的共享存储集群系统研究
引用本文:黄蓉,俞建新.基于智能存储设备的共享存储集群系统研究[J].计算机工程与设计,2007,28(16):3928-3931.
作者姓名:黄蓉  俞建新
作者单位:南京大学,计算机科学与技术系,江苏,南京,210093
摘    要:随着越来越多的机构采用集群计算技术来实现高性能计算(HPC)--地球、海洋和大气科学、地震数据分析等科学研究和药物研究、汽车设计模型、商业业务冒险分析等商业应用,集群计算技术已经进化为开发高性能计算系统的主要的方法.所有的这些应用的计算都是公认的复杂.能够有效地管理与这些应用密切相关的数据集驱动了现在的集群计算技术的发展.采用智能存储设备(OSD)大大简化了元数据服务器的工作量,并且使得系统的管理和效率都得到了很大的提高.集中描述一个新的存储体系结构--基于智能OSD的共享存储集群计算系统.

关 键 词:共享存储  高性能计算  集群  对象存储设备  存储体系结构  智能  存储设备  共享存储  集群系统  研究  device  storage  intelligent  based  cluster  system  shared  集群计算系统  体系结构  描述  效率  的管理  工作量  元数据服务器  简化  发展
文章编号:1000-7024(2007)16-3928-04
修稿时间:2006-09-18

Research on shared storage cluster system based on intelligent storage device
HUANG Rong,YU Jian-xin.Research on shared storage cluster system based on intelligent storage device[J].Computer Engineering and Design,2007,28(16):3928-3931.
Authors:HUANG Rong  YU Jian-xin
Affiliation:Department of Computer Science and Technology, Nanjing University, Nanjing 210093, China
Abstract:Coupling with more and more organizations adopted cluster computing to implement high performance computing,which covered the scientific researches of earth,ocean,atmospheric sciences,seismic data analysis,etc.,and the commercial applications of drug discover,automobile design simulations,risk analysis of financial portfolios,etc.,cluster computing has evolved as the dominant approach for developing high performance computing systems.All of these applications are recognized for their computational complexity.It is the ability to effectively manage the datasets associated with these applications that is driving current cluster computing developments.Adopting intelligent storage device(OSD) greatly reduce the workload of metadata server,and sharply improve both effective and manage of system.A new storage architecture-shared storage cluster computing system based on intelligent OSD is focused on describing.
Keywords:shared storage  high performance computing  cluster  object storage device  storage architecture
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