An efficient hash index structure for solid state disks

Hongchan Roh; Sanghyun Park

An efficient hash index structure for solid state disks

Department of Computer Science

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

Abstract

An SSD is a data storage device which has excellent features such as fast access time, shock-resistance, and power-efficiency. However, the SSD is vulnerable to small-size random writes. Therefore, a good index structure is needed in order to completely utilize these superior features and address the drawback. This paper proposes an efficient hash index structure using a write buffer and log-based flushing policy, based on linear hashing and deferred splitting. A write buffer accumulates index records, and logs them to index buckets when flushing them. Through a simple performance analysis, the proposed index confirmed that it significantly reduces the number of small-size random writes.

Original language	English
Title of host publication	Proceedings of the 2008 International Conference on Information and Knowledge Engineering, IKE 2008
Pages	256-261
Number of pages	6
Publication status	Published - 2008
Event	2008 International Conference on Information and Knowledge Engineering, IKE 2008 - Las Vegas, NV, United States Duration: 2008 Jul 14 → 2008 Jul 17

Publication series

Name	Proceedings of the 2008 International Conference on Information and Knowledge Engineering, IKE 2008

Other

Other	2008 International Conference on Information and Knowledge Engineering, IKE 2008
Country	United States
City	Las Vegas, NV
Period	08/7/14 → 08/7/17

All Science Journal Classification (ASJC) codes

Software
Artificial Intelligence
Computer Science Applications
Information Systems

Cite this

@inproceedings{bf2f46377fb14199acecab5071fe3397,

title = "An efficient hash index structure for solid state disks",

abstract = "An SSD is a data storage device which has excellent features such as fast access time, shock-resistance, and power-efficiency. However, the SSD is vulnerable to small-size random writes. Therefore, a good index structure is needed in order to completely utilize these superior features and address the drawback. This paper proposes an efficient hash index structure using a write buffer and log-based flushing policy, based on linear hashing and deferred splitting. A write buffer accumulates index records, and logs them to index buckets when flushing them. Through a simple performance analysis, the proposed index confirmed that it significantly reduces the number of small-size random writes.",

author = "Hongchan Roh and Sanghyun Park",

note = "Copyright: Copyright 2012 Elsevier B.V., All rights reserved.; 2008 International Conference on Information and Knowledge Engineering, IKE 2008 ; Conference date: 14-07-2008 Through 17-07-2008",

year = "2008",

language = "English",

isbn = "1601320752",

series = "Proceedings of the 2008 International Conference on Information and Knowledge Engineering, IKE 2008",

pages = "256--261",

booktitle = "Proceedings of the 2008 International Conference on Information and Knowledge Engineering, IKE 2008",

}

Roh, H & Park, S 2008, An efficient hash index structure for solid state disks. in Proceedings of the 2008 International Conference on Information and Knowledge Engineering, IKE 2008. Proceedings of the 2008 International Conference on Information and Knowledge Engineering, IKE 2008, pp. 256-261, 2008 International Conference on Information and Knowledge Engineering, IKE 2008, Las Vegas, NV, United States, 08/7/14.

An efficient hash index structure for solid state disks. / Roh, Hongchan; Park, Sanghyun.

Proceedings of the 2008 International Conference on Information and Knowledge Engineering, IKE 2008. 2008. p. 256-261 (Proceedings of the 2008 International Conference on Information and Knowledge Engineering, IKE 2008).

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

TY - GEN

T1 - An efficient hash index structure for solid state disks

AU - Roh, Hongchan

AU - Park, Sanghyun

PY - 2008

Y1 - 2008

N2 - An SSD is a data storage device which has excellent features such as fast access time, shock-resistance, and power-efficiency. However, the SSD is vulnerable to small-size random writes. Therefore, a good index structure is needed in order to completely utilize these superior features and address the drawback. This paper proposes an efficient hash index structure using a write buffer and log-based flushing policy, based on linear hashing and deferred splitting. A write buffer accumulates index records, and logs them to index buckets when flushing them. Through a simple performance analysis, the proposed index confirmed that it significantly reduces the number of small-size random writes.

AB - An SSD is a data storage device which has excellent features such as fast access time, shock-resistance, and power-efficiency. However, the SSD is vulnerable to small-size random writes. Therefore, a good index structure is needed in order to completely utilize these superior features and address the drawback. This paper proposes an efficient hash index structure using a write buffer and log-based flushing policy, based on linear hashing and deferred splitting. A write buffer accumulates index records, and logs them to index buckets when flushing them. Through a simple performance analysis, the proposed index confirmed that it significantly reduces the number of small-size random writes.

UR - http://www.scopus.com/inward/record.url?scp=62749095957&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=62749095957&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:62749095957

SN - 1601320752

SN - 9781601320759

T3 - Proceedings of the 2008 International Conference on Information and Knowledge Engineering, IKE 2008

SP - 256

EP - 261

BT - Proceedings of the 2008 International Conference on Information and Knowledge Engineering, IKE 2008

T2 - 2008 International Conference on Information and Knowledge Engineering, IKE 2008

Y2 - 14 July 2008 through 17 July 2008

ER -