Serverless network file systems
ACM Transactions on Computer Systems (TOCS) - Special issue on operating system principles
Frangipani: a scalable distributed file system
Proceedings of the sixteenth ACM symposium on Operating systems principles
Online computation and competitive analysis
Online computation and competitive analysis
A large-scale study of file-system contents
SIGMETRICS '99 Proceedings of the 1999 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Approximation algorithms for min-sum p-clustering
Discrete Applied Mathematics
A prototype implementation of archival Intermemory
Proceedings of the fourth ACM conference on Digital libraries
Feasibility of a serverless distributed file system deployed on an existing set of desktop PCs
Proceedings of the 2000 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
OceanStore: an architecture for global-scale persistent storage
ACM SIGPLAN Notices
Large-Scale Simulation of Replica Placement Algorithms for a Serverless Distributed File System
MASCOTS '01 Proceedings of the Ninth International Symposium in Modeling, Analysis and Simulation of Computer and Telecommunication Systems
Minimal replication cost for availability
Proceedings of the twenty-first annual symposium on Principles of distributed computing
Abstracting remote object interaction in a peer-2-peer environment
JGI '02 Proceedings of the 2002 joint ACM-ISCOPE conference on Java Grande
ESA '01 Proceedings of the 9th Annual European Symposium on Algorithms
The costs and limits of availability for replicated services
ACM Transactions on Computer Systems (TOCS)
Load balancing in dynamic structured peer-to-peer systems
Performance Evaluation - P2P computing systems
The Farsite project: a retrospective
ACM SIGOPS Operating Systems Review - Systems work at Microsoft Research
Chain replication for supporting high throughput and availability
OSDI'04 Proceedings of the 6th conference on Symposium on Opearting Systems Design & Implementation - Volume 6
Availability of multi-object operations
NSDI'06 Proceedings of the 3rd conference on Networked Systems Design & Implementation - Volume 3
Optimal inter-object correlation when replicating for availability
Proceedings of the twenty-sixth annual ACM symposium on Principles of distributed computing
Replication degree customization for high availability
Proceedings of the 3rd ACM SIGOPS/EuroSys European Conference on Computer Systems 2008
DADC '08 Proceedings of the 2008 international workshop on Data-aware distributed computing
ATTEST: ATTributes-based Extendable STorage
Journal of Systems and Software
Topology-aware replica placement in fault-tolerant embedded networks
ARCS'08 Proceedings of the 21st international conference on Architecture of computing systems
Wide area placement of data replicas for fast and highly available data access
Proceedings of the fourth international workshop on Data-intensive distributed computing
A survey of approximation results for local search algorithms
Efficient Approximation and Online Algorithms
Supporting Social Networks With Agent-Based Services
International Journal of Virtual Communities and Social Networking
Hi-index | 0.00 |
The Farsite distributed file system stores multiple replicas of files on multiple machines, to provide file access even when some machines are unavailable. Farsite assigns file replicas to machines so as to maximallyexploit the different degrees of availabilityof different machines, given an allowable replication factor R. We use competitive analysis and simulation to study the performance of three candidate hillclimbing replica placement strategies, MinMax, MinRand, and RandRand, each of which successivelyexc hanges the locations of two file replicas. We show that the MinRand and RandRand strategies are perfectlycomp etitive for R = 2 and 2/3-competitive for R = 3. For general R, MinRand is at least 1/2-competitive and RandRand is at least 10/17-competitive. The MinMax strategyis not competitive. Simulation results show better performance than the theoretic worst-case bounds.