A self-balancing striping scheme for NAND-flash storage systems

Yu Bin Chang; Li-Pin Chang

doi:10.1145/1363686.1364094

A self-balancing striping scheme for NAND-flash storage systems

Yu Bin Chang^*, Li-Pin Chang

^*Corresponding author for this work

Department of Computer Science

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

17 Scopus citations

Abstract

To use multiple memory banks in parallel is a nature approach to boost the performance of flash-memory storage systems. However, realistic data-access localities unevenly load each memory bank and thus the benefits of parallelism is severely limited. In this work, we propose to encode popular data with redundancy by means of erasure codes. Load balancing is thus achieved by accessing only lightly loaded banks, because to retrieve a subset of data blocks and code blocks sufficiently reconstructs the requested data. The technical issues pertain to redundancy allocation, redundancy placement, and request scheduling. By experiments, we found that, by offering 10% extra redundant space, the read response time is largely improved by 30%.

Original language	English
Title of host publication	Proceedings of the 23rd Annual ACM Symposium on Applied Computing, SAC'08
Pages	1715-1719
Number of pages	5
DOIs	https://doi.org/10.1145/1363686.1364094
State	Published - 1 Dec 2008
Event	23rd Annual ACM Symposium on Applied Computing, SAC'08 - Fortaleza, Ceara, Brazil Duration: 16 Mar 2008 → 20 Mar 2008

Publication series

Name	Proceedings of the ACM Symposium on Applied Computing

Conference

Conference	23rd Annual ACM Symposium on Applied Computing, SAC'08
Country	Brazil
City	Fortaleza, Ceara
Period	16/03/08 → 20/03/08

Keywords

Embedded systems
Flash memory
Storage systems

Access to Document

10.1145/1363686.1364094

Fingerprint Dive into the research topics of 'A self-balancing striping scheme for NAND-flash storage systems'. Together they form a unique fingerprint.

Cite this

@inproceedings{023680e446ef4086a8b226e550c2dce7,

title = "A self-balancing striping scheme for NAND-flash storage systems",

abstract = "To use multiple memory banks in parallel is a nature approach to boost the performance of flash-memory storage systems. However, realistic data-access localities unevenly load each memory bank and thus the benefits of parallelism is severely limited. In this work, we propose to encode popular data with redundancy by means of erasure codes. Load balancing is thus achieved by accessing only lightly loaded banks, because to retrieve a subset of data blocks and code blocks sufficiently reconstructs the requested data. The technical issues pertain to redundancy allocation, redundancy placement, and request scheduling. By experiments, we found that, by offering 10% extra redundant space, the read response time is largely improved by 30%.",

keywords = "Embedded systems, Flash memory, Storage systems",

author = "Chang, {Yu Bin} and Li-Pin Chang",

year = "2008",

month = dec,

day = "1",

doi = "10.1145/1363686.1364094",

language = "English",

isbn = "9781595937537",

series = "Proceedings of the ACM Symposium on Applied Computing",

pages = "1715--1719",

booktitle = "Proceedings of the 23rd Annual ACM Symposium on Applied Computing, SAC'08",

note = "null ; Conference date: 16-03-2008 Through 20-03-2008",

}

Chang, YB & Chang, L-P 2008, A self-balancing striping scheme for NAND-flash storage systems. in Proceedings of the 23rd Annual ACM Symposium on Applied Computing, SAC'08. Proceedings of the ACM Symposium on Applied Computing, pp. 1715-1719, 23rd Annual ACM Symposium on Applied Computing, SAC'08, Fortaleza, Ceara, Brazil, 16/03/08. https://doi.org/10.1145/1363686.1364094

TY - GEN

T1 - A self-balancing striping scheme for NAND-flash storage systems

AU - Chang, Yu Bin

AU - Chang, Li-Pin

PY - 2008/12/1

Y1 - 2008/12/1

N2 - To use multiple memory banks in parallel is a nature approach to boost the performance of flash-memory storage systems. However, realistic data-access localities unevenly load each memory bank and thus the benefits of parallelism is severely limited. In this work, we propose to encode popular data with redundancy by means of erasure codes. Load balancing is thus achieved by accessing only lightly loaded banks, because to retrieve a subset of data blocks and code blocks sufficiently reconstructs the requested data. The technical issues pertain to redundancy allocation, redundancy placement, and request scheduling. By experiments, we found that, by offering 10% extra redundant space, the read response time is largely improved by 30%.

AB - To use multiple memory banks in parallel is a nature approach to boost the performance of flash-memory storage systems. However, realistic data-access localities unevenly load each memory bank and thus the benefits of parallelism is severely limited. In this work, we propose to encode popular data with redundancy by means of erasure codes. Load balancing is thus achieved by accessing only lightly loaded banks, because to retrieve a subset of data blocks and code blocks sufficiently reconstructs the requested data. The technical issues pertain to redundancy allocation, redundancy placement, and request scheduling. By experiments, we found that, by offering 10% extra redundant space, the read response time is largely improved by 30%.

KW - Embedded systems

KW - Flash memory

KW - Storage systems

UR - http://www.scopus.com/inward/record.url?scp=56749179724&partnerID=8YFLogxK

U2 - 10.1145/1363686.1364094

DO - 10.1145/1363686.1364094

M3 - Conference contribution

AN - SCOPUS:56749179724

SN - 9781595937537

T3 - Proceedings of the ACM Symposium on Applied Computing

SP - 1715

EP - 1719

BT - Proceedings of the 23rd Annual ACM Symposium on Applied Computing, SAC'08

Y2 - 16 March 2008 through 20 March 2008

ER -