Evolving a retention period classifier for use with flash memory

Damien Hogan; Tom Arbuckle; Conor Ryan; Joe Sullivan

doi:10.5220/0004116200240033

Evolving a retention period classifier for use with flash memory

Damien Hogan, Tom Arbuckle, Conor Ryan, Joe Sullivan

Department of Electrical and Electronic Engineering

University of Limerick

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Citations (Scopus)

Abstract

Flash memory based Solid State Drives (SSDs) are gaining momentum toward replacing traditional Hard Disk Drives (HDDs) in computers and are now also generating commercial interest from enterprise data storage companies. However, storage locations in Flash memory devices degrade through repeated programming and erasing. As the storage blocks within a Flash device deteriorate through use, their ability to retain data while powered off over long periods also diminishes. Currently there is no way to predict whether a block will successfully retain data for a specified period of time while powered off. We detail our use of Genetic Programming (GP) to evolve a binary classifier which predicts whether blocks within a Flash memory device will still satisfactorily retain data after prolonged use, saving considerable amounts of testing time. This is the first time a solution to this problem has been proposed and results show an average of over 85% correct classification on previously unseen data.

Original language	English
Title of host publication	IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence
Editors	Agostinho C. Rosa, António Dourado Correia, Kurosh Madani, Joaquim Filipe, Janusz Kacprzyk
Publisher	Science and Technology Publications, Lda
Pages	24-33
Number of pages	10
ISBN (Print)	9789898565334
DOIs	https://doi.org/10.5220/0004116200240033
Publication status	Published - 2012
Event	4th International Joint Conference on Computational Intelligence, ECTA 2012 - Barcelona, Spain Duration: 5 Oct 2012 → 7 Oct 2012

Publication series

Name	IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence

Conference

Conference	4th International Joint Conference on Computational Intelligence, ECTA 2012
Country/Territory	Spain
City	Barcelona
Period	5/10/12 → 7/10/12

Keywords

Binary Classifier
Flash Memory
Genetic Programming
Solid State Drive

Access to Document

10.5220/0004116200240033

Cite this

Hogan, D., Arbuckle, T., Ryan, C., & Sullivan, J. (2012). Evolving a retention period classifier for use with flash memory. In A. C. Rosa, A. D. Correia, K. Madani, J. Filipe, & J. Kacprzyk (Eds.), IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence (pp. 24-33). (IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence). Science and Technology Publications, Lda. https://doi.org/10.5220/0004116200240033

Hogan, Damien ; Arbuckle, Tom ; Ryan, Conor et al. / Evolving a retention period classifier for use with flash memory. IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence. editor / Agostinho C. Rosa ; António Dourado Correia ; Kurosh Madani ; Joaquim Filipe ; Janusz Kacprzyk. Science and Technology Publications, Lda, 2012. pp. 24-33 (IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence).

@inproceedings{96aab54783c4443cbcf0d1bbdf609648,

title = "Evolving a retention period classifier for use with flash memory",

abstract = "Flash memory based Solid State Drives (SSDs) are gaining momentum toward replacing traditional Hard Disk Drives (HDDs) in computers and are now also generating commercial interest from enterprise data storage companies. However, storage locations in Flash memory devices degrade through repeated programming and erasing. As the storage blocks within a Flash device deteriorate through use, their ability to retain data while powered off over long periods also diminishes. Currently there is no way to predict whether a block will successfully retain data for a specified period of time while powered off. We detail our use of Genetic Programming (GP) to evolve a binary classifier which predicts whether blocks within a Flash memory device will still satisfactorily retain data after prolonged use, saving considerable amounts of testing time. This is the first time a solution to this problem has been proposed and results show an average of over 85% correct classification on previously unseen data.",

keywords = "Binary Classifier, Flash Memory, Genetic Programming, Solid State Drive",

author = "Damien Hogan and Tom Arbuckle and Conor Ryan and Joe Sullivan",

note = "Publisher Copyright: {\textcopyright} 2012 SCITEPRESS (Science and Technology Publications, Lda.).; 4th International Joint Conference on Computational Intelligence, ECTA 2012 ; Conference date: 05-10-2012 Through 07-10-2012",

year = "2012",

doi = "10.5220/0004116200240033",

language = "English",

isbn = "9789898565334",

series = "IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence",

publisher = "Science and Technology Publications, Lda",

pages = "24--33",

editor = "Rosa, {Agostinho C.} and Correia, {Ant{\'o}nio Dourado} and Kurosh Madani and Joaquim Filipe and Janusz Kacprzyk",

booktitle = "IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence",

address = "Portugal",

}

Hogan, D, Arbuckle, T, Ryan, C & Sullivan, J 2012, Evolving a retention period classifier for use with flash memory. in AC Rosa, AD Correia, K Madani, J Filipe & J Kacprzyk (eds), IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence. IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence, Science and Technology Publications, Lda, pp. 24-33, 4th International Joint Conference on Computational Intelligence, ECTA 2012, Barcelona, Spain, 5/10/12. https://doi.org/10.5220/0004116200240033

Evolving a retention period classifier for use with flash memory. / Hogan, Damien; Arbuckle, Tom; Ryan, Conor et al.
IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence. ed. / Agostinho C. Rosa; António Dourado Correia; Kurosh Madani; Joaquim Filipe; Janusz Kacprzyk. Science and Technology Publications, Lda, 2012. p. 24-33 (IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Evolving a retention period classifier for use with flash memory

AU - Hogan, Damien

AU - Arbuckle, Tom

AU - Ryan, Conor

AU - Sullivan, Joe

PY - 2012

Y1 - 2012

N2 - Flash memory based Solid State Drives (SSDs) are gaining momentum toward replacing traditional Hard Disk Drives (HDDs) in computers and are now also generating commercial interest from enterprise data storage companies. However, storage locations in Flash memory devices degrade through repeated programming and erasing. As the storage blocks within a Flash device deteriorate through use, their ability to retain data while powered off over long periods also diminishes. Currently there is no way to predict whether a block will successfully retain data for a specified period of time while powered off. We detail our use of Genetic Programming (GP) to evolve a binary classifier which predicts whether blocks within a Flash memory device will still satisfactorily retain data after prolonged use, saving considerable amounts of testing time. This is the first time a solution to this problem has been proposed and results show an average of over 85% correct classification on previously unseen data.

AB - Flash memory based Solid State Drives (SSDs) are gaining momentum toward replacing traditional Hard Disk Drives (HDDs) in computers and are now also generating commercial interest from enterprise data storage companies. However, storage locations in Flash memory devices degrade through repeated programming and erasing. As the storage blocks within a Flash device deteriorate through use, their ability to retain data while powered off over long periods also diminishes. Currently there is no way to predict whether a block will successfully retain data for a specified period of time while powered off. We detail our use of Genetic Programming (GP) to evolve a binary classifier which predicts whether blocks within a Flash memory device will still satisfactorily retain data after prolonged use, saving considerable amounts of testing time. This is the first time a solution to this problem has been proposed and results show an average of over 85% correct classification on previously unseen data.

KW - Binary Classifier

KW - Flash Memory

KW - Genetic Programming

KW - Solid State Drive

UR - http://www.scopus.com/inward/record.url?scp=84886931317&partnerID=8YFLogxK

U2 - 10.5220/0004116200240033

DO - 10.5220/0004116200240033

M3 - Conference contribution

AN - SCOPUS:84886931317

SN - 9789898565334

T3 - IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence

SP - 24

EP - 33

BT - IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence

A2 - Rosa, Agostinho C.

A2 - Correia, António Dourado

A2 - Madani, Kurosh

A2 - Filipe, Joaquim

A2 - Kacprzyk, Janusz

PB - Science and Technology Publications, Lda

T2 - 4th International Joint Conference on Computational Intelligence, ECTA 2012

Y2 - 5 October 2012 through 7 October 2012

ER -

Hogan D, Arbuckle T, Ryan C, Sullivan J. Evolving a retention period classifier for use with flash memory. In Rosa AC, Correia AD, Madani K, Filipe J, Kacprzyk J, editors, IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence. Science and Technology Publications, Lda. 2012. p. 24-33. (IJCCI 2012 - Proceedings of the 4th International Joint Conference on Computational Intelligence). doi: 10.5220/0004116200240033

Evolving a retention period classifier for use with flash memory

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this