Dynamic Metadata Management for Petabyte-Scale File Systems
Top Cited Papers
- 1 April 2005
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
In petabyte-scale distributed file systems that decouple read and write from metadata operations, behavior of the metadata server cluster will be critical to overall system performance and scalability. We present a dynamic subtree partitioning and adaptive metadata management system designed to efficiently manage hierarchical metadata workloads that evolve over time. We examine the relative merits of our approach in the context of traditional workload partitioning strategies, and demonstrate the performance, scalability and adaptability advantages in a simulation environment.Keywords
This publication has 14 references indexed in Scilit:
- Replication under scalable hashing:a family of algorithms for scalable decentralized data distributionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2004
- Reliability mechanisms for very large storage systemsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- zFS - a scalable distributed file system using object disksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Efficient metadata management in large distributed storage systemsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- RAMA: An easy-to-use, high-performance parallel file systemParallel Computing, 1997
- Techniques for file system simulationSoftware: Practice and Experience, 1994
- Coda: a highly available file system for a distributed workstation environmentIEEE Transactions on Computers, 1990
- The Sprite network operating systemComputer, 1988
- Andrew: a distributed personal computing environmentCommunications of the ACM, 1986
- A fast file system for UNIXACM Transactions on Computer Systems, 1984