Distributed Storage Layout Schemes

Peter Sobe

doi:10.1109/IPDPS.2005.172

Distributed Storage Layout Schemes

Peter Sobe

Institute of Computer Engineering

Abstract

When distributing data across several nodes, two different approaches exist. The first one consists in distribution of the data object itself, e.g. in striping. The second approach is aggregation of local storages, whereby each data object is assigned to a home storage node. From the viewpoint of fault-tolerant data layouts, these schemes seem to be similar. In both cases the addition of parity, e.g. RAID level 3, level 5 or Reed-Solomon codes provide tolerance against node failures. A closer look shows differences in reachable access rates, needed number of messages and recovery cost. In this paper we compare both approaches and provide a method for self reconfiguration. The transformation from a parity grouping layout to a striping layout is shown to be feasible for stepwise and concurrent operation during data access.

Original language	English
Title of host publication	19th IEEE International Parallel and Distributed Processing Symposium
Publisher	IEEE
Publication date	01.12.2005
Article number	1420248
ISBN (Print)	0-7695-2312-9, 978-076952312-5
DOIs	https://doi.org/10.1109/IPDPS.2005.172
Publication status	Published - 01.12.2005
Event	19th IEEE International Parallel and Distributed Processing Symposium - Denver , United States Duration: 04.04.2005 → 08.04.2005 Conference number: 67370

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/IPDPS.2005.172

Cite this

@inproceedings{a03f4e6440a540c59af2b26302420193,

title = "Distributed Storage Layout Schemes",

abstract = "When distributing data across several nodes, two different approaches exist. The first one consists in distribution of the data object itself, e.g. in striping. The second approach is aggregation of local storages, whereby each data object is assigned to a home storage node. From the viewpoint of fault-tolerant data layouts, these schemes seem to be similar. In both cases the addition of parity, e.g. RAID level 3, level 5 or Reed-Solomon codes provide tolerance against node failures. A closer look shows differences in reachable access rates, needed number of messages and recovery cost. In this paper we compare both approaches and provide a method for self reconfiguration. The transformation from a parity grouping layout to a striping layout is shown to be feasible for stepwise and concurrent operation during data access.",

author = "Peter Sobe",

year = "2005",

month = dec,

day = "1",

doi = "10.1109/IPDPS.2005.172",

language = "English",

isbn = "0-7695-2312-9",

series = "IEEE International Parallel and Distributed Processing Symposium",

publisher = "IEEE",

booktitle = "19th IEEE International Parallel and Distributed Processing Symposium",

note = "19th IEEE International Parallel and Distributed Processing Symposium , IPDPS 2005 ; Conference date: 04-04-2005 Through 08-04-2005",

}

TY - GEN

T1 - Distributed Storage Layout Schemes

AU - Sobe, Peter

N1 - Conference code: 67370

PY - 2005/12/1

Y1 - 2005/12/1

N2 - When distributing data across several nodes, two different approaches exist. The first one consists in distribution of the data object itself, e.g. in striping. The second approach is aggregation of local storages, whereby each data object is assigned to a home storage node. From the viewpoint of fault-tolerant data layouts, these schemes seem to be similar. In both cases the addition of parity, e.g. RAID level 3, level 5 or Reed-Solomon codes provide tolerance against node failures. A closer look shows differences in reachable access rates, needed number of messages and recovery cost. In this paper we compare both approaches and provide a method for self reconfiguration. The transformation from a parity grouping layout to a striping layout is shown to be feasible for stepwise and concurrent operation during data access.

AB - When distributing data across several nodes, two different approaches exist. The first one consists in distribution of the data object itself, e.g. in striping. The second approach is aggregation of local storages, whereby each data object is assigned to a home storage node. From the viewpoint of fault-tolerant data layouts, these schemes seem to be similar. In both cases the addition of parity, e.g. RAID level 3, level 5 or Reed-Solomon codes provide tolerance against node failures. A closer look shows differences in reachable access rates, needed number of messages and recovery cost. In this paper we compare both approaches and provide a method for self reconfiguration. The transformation from a parity grouping layout to a striping layout is shown to be feasible for stepwise and concurrent operation during data access.

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-33746289578&origin=inward&txGid=ddca48724e5bbb2e8b14cc67482fd3ff

U2 - 10.1109/IPDPS.2005.172

DO - 10.1109/IPDPS.2005.172

M3 - Conference contribution

SN - 0-7695-2312-9

SN - 978-076952312-5

T3 - IEEE International Parallel and Distributed Processing Symposium

BT - 19th IEEE International Parallel and Distributed Processing Symposium

PB - IEEE

T2 - 19th IEEE International Parallel and Distributed Processing Symposium

Y2 - 4 April 2005 through 8 April 2005

ER -

Distributed Storage Layout Schemes

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this