Design and Implementation of Distributed Memory Management System Based on Local Area Network

被引:0
作者
Wang Yujian [1 ]
Fan Lili [1 ]
Liu Chen [2 ]
Tian Xing [2 ]
机构
[1] Beijing Union Univ, Inst Informat Technol, Beijing, Peoples R China
[2] Beijing Univ Posts & Telecommun, Coll Comp Sci & Technol, Beijing 100876, Peoples R China
来源
FIRST INTERNATIONAL WORKSHOP ON DATABASE TECHNOLOGY AND APPLICATIONS, PROCEEDINGS | 2009年
关键词
distributed computing; memory management; package; open shortest path first (OSPF);
D O I
10.1109/DBTA.2009.162
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
According to the characteristics of data processing in the middle-small enterprises, a distributed memory management system (DMMS) based on LAN is proposed. The manner of data organizing, the design principle and the concrete implementation of every function entity in the system are introduced in detail. The system uses the memory resource of the computers in the LAN to store user application data, solves the problem about I/O capability bottleneck. The data is divided into the packages, and the package management entity (PME) is used for the management of network memory resources. Main package is kept synchronized with backup package by synchronization protocol mechanism to realize the protection of package data. The whole system design adopts flexible framework, each function entity is relatively independent of each other and the protocols and algorithms used in the system are encapsulated to realize agility and expansibility. The results of test show that the efficiency of processing application data by using the system is 5 to 7 times as high as that by using the local fixed disk for user, it is really feasible and effective to apply the distributed technology in the memory management of LAN.
引用
收藏
页码:83 / +
页数:2
相关论文
共 8 条
[1]  
Abadi Daniel, 2006, P 2006 ACM SIGMOD IN, P671, DOI [DOI 10.1145/1142473.1142548, 10.1145/1142473.1142548]
[2]  
[Anonymous], 1998, 2328 RFC
[3]   Modeling and analysis of using memory management unit to improve software reliability [J].
Chang, SJ ;
Kapauan, PTZ .
12TH INTERNATIONAL SYMPOSIUM ON SOFTWARE RELIABILITY ENGINEERING, PROCEEDINGS, 2001, :96-102
[4]  
DAN LC, 2004, J SYST SOFTWARE, V72, P443
[5]   Fast and secure distributed read-only file system [J].
Fu, K ;
Kaashoek, MF ;
Mazières, D .
ACM TRANSACTIONS ON COMPUTER SYSTEMS, 2002, 20 (01) :1-24
[6]   An optimal scheduling algorithm based on task duplication [J].
Park, CK ;
Choe, TY .
IEEE TRANSACTIONS ON COMPUTERS, 2002, 51 (04) :444-448
[7]  
Wang Shan, 2007, Journal of Computer Applications, V27, P2353
[8]  
Xiao Ying-yuan, 2005, Wuhan University Journal of Natural Sciences, V10, P61, DOI 10.1007/BF02828618