A framework for efficient evaluation of the fault tolerance of deduplicated storage systems

Eric William Davis Rozier, William H. Sanders

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

8 Citations (Scopus)

Abstract

In this paper we present a framework for analyzing the fault tolerance of deduplicated storage systems. We discuss methods for building models of deduplicated storage systems by analyzing empirical data on a file category basis. We provide an algorithm for generating component-based models from this information and a specification of the storage system architecture. Given the complex nature of detailed models of deduplicated storage systems, finding a solution using traditional discrete event simulation or numerical solvers can be difficult. We introduce an algorithm which allows for a more efficient solution by exploiting the underlying structure of dependencies to decompose the model of the storage system. We present a case study of our framework for a real system.We analyze a production deduplicated storage system and propose extensions which improve fault tolerance while maintaining high storage efficiency.

Original languageEnglish
Title of host publicationProceedings of the International Conference on Dependable Systems and Networks
DOIs
StatePublished - Oct 1 2012
Event42nd Annual IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2012 - Boston, MA, United States
Duration: Jun 25 2012Jun 28 2012

Other

Other42nd Annual IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2012
CountryUnited States
CityBoston, MA
Period6/25/126/28/12

Fingerprint

Fault tolerance
Discrete event simulation
Specifications

Keywords

  • decomposition
  • deduplication
  • reliability
  • simulation
  • storage

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Hardware and Architecture
  • Software

Cite this

Rozier, E. W. D., & Sanders, W. H. (2012). A framework for efficient evaluation of the fault tolerance of deduplicated storage systems. In Proceedings of the International Conference on Dependable Systems and Networks [6263921] https://doi.org/10.1109/DSN.2012.6263921

A framework for efficient evaluation of the fault tolerance of deduplicated storage systems. / Rozier, Eric William Davis; Sanders, William H.

Proceedings of the International Conference on Dependable Systems and Networks. 2012. 6263921.

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

Rozier, EWD & Sanders, WH 2012, A framework for efficient evaluation of the fault tolerance of deduplicated storage systems. in Proceedings of the International Conference on Dependable Systems and Networks., 6263921, 42nd Annual IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2012, Boston, MA, United States, 6/25/12. https://doi.org/10.1109/DSN.2012.6263921
Rozier EWD, Sanders WH. A framework for efficient evaluation of the fault tolerance of deduplicated storage systems. In Proceedings of the International Conference on Dependable Systems and Networks. 2012. 6263921 https://doi.org/10.1109/DSN.2012.6263921
Rozier, Eric William Davis ; Sanders, William H. / A framework for efficient evaluation of the fault tolerance of deduplicated storage systems. Proceedings of the International Conference on Dependable Systems and Networks. 2012.
@inproceedings{496438ffb03e4a12a09e91c2de4a3e59,
title = "A framework for efficient evaluation of the fault tolerance of deduplicated storage systems",
abstract = "In this paper we present a framework for analyzing the fault tolerance of deduplicated storage systems. We discuss methods for building models of deduplicated storage systems by analyzing empirical data on a file category basis. We provide an algorithm for generating component-based models from this information and a specification of the storage system architecture. Given the complex nature of detailed models of deduplicated storage systems, finding a solution using traditional discrete event simulation or numerical solvers can be difficult. We introduce an algorithm which allows for a more efficient solution by exploiting the underlying structure of dependencies to decompose the model of the storage system. We present a case study of our framework for a real system.We analyze a production deduplicated storage system and propose extensions which improve fault tolerance while maintaining high storage efficiency.",
keywords = "decomposition, deduplication, reliability, simulation, storage",
author = "Rozier, {Eric William Davis} and Sanders, {William H.}",
year = "2012",
month = "10",
day = "1",
doi = "10.1109/DSN.2012.6263921",
language = "English",
isbn = "9781467316248",
booktitle = "Proceedings of the International Conference on Dependable Systems and Networks",

}

TY - GEN

T1 - A framework for efficient evaluation of the fault tolerance of deduplicated storage systems

AU - Rozier, Eric William Davis

AU - Sanders, William H.

PY - 2012/10/1

Y1 - 2012/10/1

N2 - In this paper we present a framework for analyzing the fault tolerance of deduplicated storage systems. We discuss methods for building models of deduplicated storage systems by analyzing empirical data on a file category basis. We provide an algorithm for generating component-based models from this information and a specification of the storage system architecture. Given the complex nature of detailed models of deduplicated storage systems, finding a solution using traditional discrete event simulation or numerical solvers can be difficult. We introduce an algorithm which allows for a more efficient solution by exploiting the underlying structure of dependencies to decompose the model of the storage system. We present a case study of our framework for a real system.We analyze a production deduplicated storage system and propose extensions which improve fault tolerance while maintaining high storage efficiency.

AB - In this paper we present a framework for analyzing the fault tolerance of deduplicated storage systems. We discuss methods for building models of deduplicated storage systems by analyzing empirical data on a file category basis. We provide an algorithm for generating component-based models from this information and a specification of the storage system architecture. Given the complex nature of detailed models of deduplicated storage systems, finding a solution using traditional discrete event simulation or numerical solvers can be difficult. We introduce an algorithm which allows for a more efficient solution by exploiting the underlying structure of dependencies to decompose the model of the storage system. We present a case study of our framework for a real system.We analyze a production deduplicated storage system and propose extensions which improve fault tolerance while maintaining high storage efficiency.

KW - decomposition

KW - deduplication

KW - reliability

KW - simulation

KW - storage

UR - http://www.scopus.com/inward/record.url?scp=84866701484&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866701484&partnerID=8YFLogxK

U2 - 10.1109/DSN.2012.6263921

DO - 10.1109/DSN.2012.6263921

M3 - Conference contribution

SN - 9781467316248

BT - Proceedings of the International Conference on Dependable Systems and Networks

ER -