Harmonized memory system for object-based cloud storage

Su Kyung Yoon, Young Sun Youn, Min Ho Son, Shin-Dug Kim

Research output: Contribution to journalArticle

Abstract

A new storage system that integrates non-volatile with conventional memory, a harmonized memory system (HMS) for object-based cloud storage, is proposed. The system overcomes IO bottlenecks when managing large amounts of metadata and transaction logs and is composed of five modules. The first, the harmonized memory supervisor, is a translation layer for accessing the harmonized array module. It manages address translation, address mapping by page linking, and wear leveling. The second, the harmonized array module, is divided into dynamic and static areas composed of DRAM, and PCM together with NAND flash memory, respectively. The harmonized memory migration engine and data pattern predictor, which anticipates future data flow, are designed to maximize the effectiveness of the PCM array area. The harmonized logging conductor processes the log between the PCMarray andNANDflash areas. Experimental results show the total execution time and energy consumption of HMS is 5.77 faster and 4.27 times lower, respectively, than the conventional DRAM-HDD model for object-based storage workloads.

Original languageEnglish
JournalCluster Computing
Volume21
Issue number1
DOIs
Publication statusPublished - 2017 May 3

Fingerprint

Data storage equipment
Pulse code modulation
Dynamic random access storage
Flash memory
Supervisory personnel
Metadata
Computer systems
Energy utilization
Wear of materials
Engines

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Networks and Communications

Cite this

Yoon, Su Kyung ; Youn, Young Sun ; Son, Min Ho ; Kim, Shin-Dug. / Harmonized memory system for object-based cloud storage. In: Cluster Computing. 2017 ; Vol. 21, No. 1.
@article{295b8c61a4674fd8b3b8eb9fad938fcd,
title = "Harmonized memory system for object-based cloud storage",
abstract = "A new storage system that integrates non-volatile with conventional memory, a harmonized memory system (HMS) for object-based cloud storage, is proposed. The system overcomes IO bottlenecks when managing large amounts of metadata and transaction logs and is composed of five modules. The first, the harmonized memory supervisor, is a translation layer for accessing the harmonized array module. It manages address translation, address mapping by page linking, and wear leveling. The second, the harmonized array module, is divided into dynamic and static areas composed of DRAM, and PCM together with NAND flash memory, respectively. The harmonized memory migration engine and data pattern predictor, which anticipates future data flow, are designed to maximize the effectiveness of the PCM array area. The harmonized logging conductor processes the log between the PCMarray andNANDflash areas. Experimental results show the total execution time and energy consumption of HMS is 5.77 faster and 4.27 times lower, respectively, than the conventional DRAM-HDD model for object-based storage workloads.",
author = "Yoon, {Su Kyung} and Youn, {Young Sun} and Son, {Min Ho} and Shin-Dug Kim",
year = "2017",
month = "5",
day = "3",
doi = "10.1007/s10586-017-0904-6",
language = "English",
volume = "21",
journal = "Cluster Computing",
issn = "1386-7857",
publisher = "Kluwer Academic Publishers",
number = "1",

}

Harmonized memory system for object-based cloud storage. / Yoon, Su Kyung; Youn, Young Sun; Son, Min Ho; Kim, Shin-Dug.

In: Cluster Computing, Vol. 21, No. 1, 03.05.2017.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Harmonized memory system for object-based cloud storage

AU - Yoon, Su Kyung

AU - Youn, Young Sun

AU - Son, Min Ho

AU - Kim, Shin-Dug

PY - 2017/5/3

Y1 - 2017/5/3

N2 - A new storage system that integrates non-volatile with conventional memory, a harmonized memory system (HMS) for object-based cloud storage, is proposed. The system overcomes IO bottlenecks when managing large amounts of metadata and transaction logs and is composed of five modules. The first, the harmonized memory supervisor, is a translation layer for accessing the harmonized array module. It manages address translation, address mapping by page linking, and wear leveling. The second, the harmonized array module, is divided into dynamic and static areas composed of DRAM, and PCM together with NAND flash memory, respectively. The harmonized memory migration engine and data pattern predictor, which anticipates future data flow, are designed to maximize the effectiveness of the PCM array area. The harmonized logging conductor processes the log between the PCMarray andNANDflash areas. Experimental results show the total execution time and energy consumption of HMS is 5.77 faster and 4.27 times lower, respectively, than the conventional DRAM-HDD model for object-based storage workloads.

AB - A new storage system that integrates non-volatile with conventional memory, a harmonized memory system (HMS) for object-based cloud storage, is proposed. The system overcomes IO bottlenecks when managing large amounts of metadata and transaction logs and is composed of five modules. The first, the harmonized memory supervisor, is a translation layer for accessing the harmonized array module. It manages address translation, address mapping by page linking, and wear leveling. The second, the harmonized array module, is divided into dynamic and static areas composed of DRAM, and PCM together with NAND flash memory, respectively. The harmonized memory migration engine and data pattern predictor, which anticipates future data flow, are designed to maximize the effectiveness of the PCM array area. The harmonized logging conductor processes the log between the PCMarray andNANDflash areas. Experimental results show the total execution time and energy consumption of HMS is 5.77 faster and 4.27 times lower, respectively, than the conventional DRAM-HDD model for object-based storage workloads.

UR - http://www.scopus.com/inward/record.url?scp=85046666703&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046666703&partnerID=8YFLogxK

U2 - 10.1007/s10586-017-0904-6

DO - 10.1007/s10586-017-0904-6

M3 - Article

VL - 21

JO - Cluster Computing

JF - Cluster Computing

SN - 1386-7857

IS - 1

ER -