The MOSIX Direct File System Access Method for Supporting Scalable Cluster File Systems

被引:18
作者
Lior Amar
Amnon Barak
Amnon Shiloh
机构
[1] The Hebrew University of Jerusalem,Institute of Computer Science
关键词
cluster computing; cluster file systems; consistent file system; massive parallel I/O;
D O I
10.1023/B:CLUS.0000018563.68085.4b
中图分类号
学科分类号
摘要
MOSIX is a cluster management system that supports preemptive process migration. This paper presents the MOSIX Direct File System Access (DFSA), a provision that can improve the performance of cluster file systems by allowing a migrated process to directly access files in its current location. This capability, when combined with an appropriate file system, could substantially increase the I/O performance and reduce the network congestion by migrating an I/O intensive process to a file server rather than the traditional way of bringing the file's data to the process. DFSA is suitable for clusters that manage a pool of shared disks among multiple machines. With DFSA, it is possible to migrate parallel processes from a client node to file servers for parallel access to different files. Any consistent file system can be adjusted to work with DFSA. To test its performance, we developed the MOSIX File-System (MFS) which allows consistent parallel operations on different files. The paper describes DFSA and presents the performance of MFS with and without DFSA.
引用
收藏
页码:141 / 150
页数:9
相关论文
共 22 条
[1]
Anderson T.E.(1996)Serverless network file systems ACM Transactions on Computer Systems 14 41-79
[2]
Dahlin M.D.(1998)Memory ushering in a scalable computing cluster Journal of Microprocessors and Microsystems 22 175-182
[3]
Neefe J.M.(1997)TheMOSIX multicomputer operating system for high performance cluster computing Future Generation Computer Systems 12 361-372
[4]
Patterson D.A.(2002)Processing large-scale multidimensional data in parallel and distributed environments Parallel Computing 28 827-859
[5]
Roselli D.S.(1997)Exploiting process lifetime distributions for dynamic load balancing ACM Transactions on Computer Systems 15 253-285
[6]
Wang R.Y.(2003)Opportunity cost algorithms for reduction of I/O and interprocess communication overhead in a computing cluster IEEE Transactions on Parallel and Distributed Systems 14 39-50
[7]
Barak A.(undefined)undefined undefined undefined undefined-undefined
[8]
Braverman A.(undefined)undefined undefined undefined undefined-undefined
[9]
Barak A.(undefined)undefined undefined undefined undefined-undefined
[10]
La O.(undefined)undefined undefined undefined undefined-undefined