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. |
Year | DOI | Venue |
---|---|---|
2004 | 10.1109/SC.2004.22 | Proceedings of the 2006 ACM/IEEE conference on Supercomputing |
Keywords | Field | DocType |
adaptive systems,performance,design,system design,computer architecture,distributed file system,languages,scalability,scientific computing,system performance,file servers,context modeling | Metadata,File server,Petabyte,Computer science,Meta Data Services,Parallel computing,Context model,Metadata management,Computer cluster,Database,Scalability,Distributed computing | Conference |
ISBN | Citations | PageRank |
0-7695-2153-3 | 104 | 4.74 |
References | Authors | |
23 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Sage A. Weil | 1 | 736 | 35.96 |
Kristal T. Pollack | 2 | 244 | 14.70 |
Scott A. Brandt | 3 | 1663 | 94.81 |
Ethan L. Miller | 4 | 2870 | 281.96 |