Efficient random network coding for distributed storage systems

Ádám Visegrádi, P. Kacsuk

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Making distributed storage systems reliable is an important challenge. Simple replication may cause severe storage overhead when individual components of The system are very unreliable. Using erasure codes is a promising solution for This problem, but it presents its own challenges; it makes The design of such a system very complex, and it presents The problem of reparation. Network coding has been suggested To be used in The communication in These networks To help reduce overhead. However, using random network coding as -not besides -erasure coding would be an even more promising field To investigate; such a system would have a simple design, need little or no centralization, and reparation of The system could be much simpler Than it is in other erasure coding schemes. The first step on This path is To investigate whether network coding can achieve such a performance That it is a feasible alternative To other erasure codes. This paper presents our experiences about The realization of random network coding based on The discrete logarithm of The finite field. We discuss possible performance optimizations for such a system, and provide performance measurement results focusing on data storage scenarios.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages385-394
Number of pages10
Volume8374 LNCS
ISBN (Print)9783642544194
DOIs
Publication statusPublished - 2014
Event19th International Conference on Parallel Processing Workshops, Euro-Par 2013 - BigDataCloud, DIHC, FedICI, HeteroPar, HiBB, LSDVE, MHPC, OMHI, PADABS, PROPER, Resilience, ROME, and UCHPC 2013 - Aachen, Germany
Duration: Aug 26 2013Aug 27 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8374 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other19th International Conference on Parallel Processing Workshops, Euro-Par 2013 - BigDataCloud, DIHC, FedICI, HeteroPar, HiBB, LSDVE, MHPC, OMHI, PADABS, PROPER, Resilience, ROME, and UCHPC 2013
CountryGermany
CityAachen
Period8/26/138/27/13

    Fingerprint

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Visegrádi, Á., & Kacsuk, P. (2014). Efficient random network coding for distributed storage systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8374 LNCS, pp. 385-394). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8374 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-642-54420-0_38