Efficient metadata management in large distributed storage systems

27 August 2003

proceedings article
Published by Institute of Electrical and Electronics Engineers (IEEE)

Vol. t10, 290-298
https://doi.org/10.1109/mass.2003.1194865

Abstract

Efficient metadata management is a critical aspect of overall system performance in large distributed storage systems. Directory subtree partitioning and pure hashing are two common techniques used for managing metadata in such systems, but both suffer from bottlenecks at very high concurrent access rates. We present a new approach called Lazy Hybrid (LH) metadata management that combines the best aspects of these two approaches while avoiding their shortcomings.

Keywords

This publication has 11 references indexed in Scilit:

Network attached storage architecture
Communications of the ACM, 2000
RAMA: An easy-to-use, high-performance parallel file system
Parallel Computing, 1997
The Vesta parallel file system
ACM Transactions on Computer Systems, 1996
Long Term Distributed File Reference Tracing: Implementation and Experience
Software: Practice and Experience, 1996
Distributed file systems: concepts and examples
ACM Computing Surveys, 1990
Coda: a highly available file system for a distributed workstation environment
IEEE Transactions on Computers, 1990
The Sprite network operating system
Computer, 1988
Andrew: a distributed personal computing environment
Communications of the ACM, 1986
A trace-driven analysis of the UNIX 4.2 BSD file system
Published by Association for Computing Machinery (ACM) ,1985
A fast file system for UNIX
ACM Transactions on Computer Systems, 1984