On the effectiveness of recoding-based repair in network coded distributed storage

Marton A. Sipos, Patrik J. Braun, Daniel Enrique Lucani Roetter, Frank Hanns Paul Fitzek, Hassan Charaf

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

2 Citationer (Scopus)
111 Downloads (Pure)

Resumé

High capacity storage systems distribute less across several storage devices (nodes) and apply an erasure code to meet availability and reliability requirements.
Since devices can lose network connectivity or fail permanently, a dynamic repair mechanism must be put in place. In such cases a new recovery node gets connected to a given subset of the operating nodes and receives a part of the stored data. The objective of this paper is to investigate data survival for Random Linear Network Coding (RLNC) as a function of topology and communication overhead, dened by the number of connections and the number of transmitted packets to the recovery node, respectively. The paper includes two main contributions. First, a sufficient set of conditions for quasi-infinite longevity of the stored data is derived. Second,
a comparison using experimental results shows that RLNC can be up to 50% more effective than traditional erasure codes like Reed-Solomon.
OriginalsprogEngelsk
TidsskriftPeriodica Polytechnica. Electrical Engineering and Computer Science
Vol/bind61
Udgave nummer1
Sider (fra-til)12-21
ISSN2064-5260
DOI
StatusUdgivet - 2017

Fingerprint

Linear networks
Network coding
Repair
Recovery
Topology
Availability
Communication

Citer dette

Sipos, Marton A. ; Braun, Patrik J. ; Roetter, Daniel Enrique Lucani ; Fitzek, Frank Hanns Paul ; Charaf, Hassan. / On the effectiveness of recoding-based repair in network coded distributed storage. I: Periodica Polytechnica. Electrical Engineering and Computer Science. 2017 ; Bind 61, Nr. 1. s. 12-21.
@article{c238965dc1db496cb8710093e14f0d75,
title = "On the effectiveness of recoding-based repair in network coded distributed storage",
abstract = "High capacity storage systems distribute less across several storage devices (nodes) and apply an erasure code to meet availability and reliability requirements.Since devices can lose network connectivity or fail permanently, a dynamic repair mechanism must be put in place. In such cases a new recovery node gets connected to a given subset of the operating nodes and receives a part of the stored data. The objective of this paper is to investigate data survival for Random Linear Network Coding (RLNC) as a function of topology and communication overhead, dened by the number of connections and the number of transmitted packets to the recovery node, respectively. The paper includes two main contributions. First, a sufficient set of conditions for quasi-infinite longevity of the stored data is derived. Second,a comparison using experimental results shows that RLNC can be up to 50{\%} more effective than traditional erasure codes like Reed-Solomon.",
author = "Sipos, {Marton A.} and Braun, {Patrik J.} and Roetter, {Daniel Enrique Lucani} and Fitzek, {Frank Hanns Paul} and Hassan Charaf",
year = "2017",
doi = "10.3311/PPee.9377",
language = "English",
volume = "61",
pages = "12--21",
journal = "Periodica Polytechnica. Electrical Engineering and Computer Science",
issn = "2064-5260",
publisher = "Budapest University of Technology and Economics",
number = "1",

}

On the effectiveness of recoding-based repair in network coded distributed storage. / Sipos, Marton A.; Braun, Patrik J. ; Roetter, Daniel Enrique Lucani; Fitzek, Frank Hanns Paul; Charaf, Hassan.

I: Periodica Polytechnica. Electrical Engineering and Computer Science, Bind 61, Nr. 1, 2017, s. 12-21.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - On the effectiveness of recoding-based repair in network coded distributed storage

AU - Sipos, Marton A.

AU - Braun, Patrik J.

AU - Roetter, Daniel Enrique Lucani

AU - Fitzek, Frank Hanns Paul

AU - Charaf, Hassan

PY - 2017

Y1 - 2017

N2 - High capacity storage systems distribute less across several storage devices (nodes) and apply an erasure code to meet availability and reliability requirements.Since devices can lose network connectivity or fail permanently, a dynamic repair mechanism must be put in place. In such cases a new recovery node gets connected to a given subset of the operating nodes and receives a part of the stored data. The objective of this paper is to investigate data survival for Random Linear Network Coding (RLNC) as a function of topology and communication overhead, dened by the number of connections and the number of transmitted packets to the recovery node, respectively. The paper includes two main contributions. First, a sufficient set of conditions for quasi-infinite longevity of the stored data is derived. Second,a comparison using experimental results shows that RLNC can be up to 50% more effective than traditional erasure codes like Reed-Solomon.

AB - High capacity storage systems distribute less across several storage devices (nodes) and apply an erasure code to meet availability and reliability requirements.Since devices can lose network connectivity or fail permanently, a dynamic repair mechanism must be put in place. In such cases a new recovery node gets connected to a given subset of the operating nodes and receives a part of the stored data. The objective of this paper is to investigate data survival for Random Linear Network Coding (RLNC) as a function of topology and communication overhead, dened by the number of connections and the number of transmitted packets to the recovery node, respectively. The paper includes two main contributions. First, a sufficient set of conditions for quasi-infinite longevity of the stored data is derived. Second,a comparison using experimental results shows that RLNC can be up to 50% more effective than traditional erasure codes like Reed-Solomon.

U2 - 10.3311/PPee.9377

DO - 10.3311/PPee.9377

M3 - Journal article

VL - 61

SP - 12

EP - 21

JO - Periodica Polytechnica. Electrical Engineering and Computer Science

JF - Periodica Polytechnica. Electrical Engineering and Computer Science

SN - 2064-5260

IS - 1

ER -