Title :
Reliability Study of Coding Schemes for Wide-Area Distributed Storage Systems
Author_Institution :
Comput. Sci. Res., Zuse Inst. Berlin (TIB), Berlin, Germany
Abstract :
Distributed storage systems comprise a large number of commodity hardware distributed across several data centers. Even in the presence of failures (permanent failures) the system should provide reliable storage. While replication has advantages because of its simplicity there exist coding techniques that provide adaptable reliability properties with an optimal redundancy ratio at the same time e.g. MDS (maximum distance separable) erasure codes. The coding and distribution scheme influences the prospective storage reliability. In this paper we present reliability models for erasure coding and replication techniques especially for their application in wide-area storage systems. Furthermore we utilize these models to quantify the reliability properties of concrete data storage scenarios.
Keywords :
computer centres; distributed processing; fault tolerant computing; forward error correction; redundancy; software reliability; storage management; system recovery; MDS erasure code; coding technique; commodity hardware; data center; maximum distance separability; optimal redundancy ratio; storage reliability; system failure; wide area distributed storage system; Data models; Encoding; Maintenance engineering; Markov processes; Redundancy; Servers; distributed data storage; erasure codes; reliability;
Conference_Titel :
Parallel, Distributed and Network-Based Processing (PDP), 2011 19th Euromicro International Conference on
Conference_Location :
Ayia Napa
Print_ISBN :
978-1-4244-9682-2
DOI :
10.1109/PDP.2011.70
