LAWC: Optimizing Write Cache Using Layout-Aware I/O Scheduling for All Flash Storage

Kalidas Ganesh, Youngjae Kim, Monobrata Debnath, Sungyong Park, Junghee Lee

Research output: Contribution to journalArticle

Abstract

Flash memory-based SSD-RAIDs are swiftly replacing conventional hard disk drives by exhibiting improved performance and stability, especially in I/O-intensive environments. However, the variations in latency and throughput occurring due to uncoordinated internal garbage collection cripples further boosting of performance. In addition, the unwanted variations in each SSD can influence the overall performance of the entire flash storage adversely. This performance bottleneck can be essentially reduced by an internal write cache in the RAID controller designed prudently by considering the crucial device characteristics. The state-of-the-art cache write for the RAID controller fails to incorporate device characteristics of flash memory-based SSDs and mitigates the performance gain. In this paper, we propose a novel cache design namely Layout-Aware Write Cache (LAWC) to overcome the performance barrier inculcated by independent garbage collections. LAWC implements (i) improved I/O scheduling for logically partitioned write caches, (ii) a destage write synchronization mechanism to allow individual write caches to flush write blocks into the SSD array in a coordinated manner, and (iii) a two-level hybrid cache algorithm utilizing small front level cache for the improved write cache efficiency. LAWC shows significant reduction in response time by 82.39 percent on RAID-0 and 68.51 percent on RAID-5 types of SSDs when compared with state-of-the-art write cache algorithms.

Original languageEnglish
Article number7932926
Pages (from-to)1890-1902
Number of pages13
JournalIEEE Transactions on Computers
Volume66
Issue number11
DOIs
Publication statusPublished - 2017 Nov 1
Externally publishedYes

Fingerprint

Flash memory
Flash
Cache
Layout
Scheduling
Controllers
Hard disk storage
Synchronization
Throughput
Garbage Collection
Flash Memory
Percent
Internal
Controller
Boosting
Response Time
Latency
Entire

Keywords

  • Flash memory
  • I/O scheduling
  • RAID
  • solid-state drive
  • storage system
  • write cache

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computational Theory and Mathematics

Cite this

LAWC : Optimizing Write Cache Using Layout-Aware I/O Scheduling for All Flash Storage. / Ganesh, Kalidas; Kim, Youngjae; Debnath, Monobrata; Park, Sungyong; Lee, Junghee.

In: IEEE Transactions on Computers, Vol. 66, No. 11, 7932926, 01.11.2017, p. 1890-1902.

Research output: Contribution to journalArticle

Ganesh, Kalidas ; Kim, Youngjae ; Debnath, Monobrata ; Park, Sungyong ; Lee, Junghee. / LAWC : Optimizing Write Cache Using Layout-Aware I/O Scheduling for All Flash Storage. In: IEEE Transactions on Computers. 2017 ; Vol. 66, No. 11. pp. 1890-1902.
@article{bfcf452f9a944049a7f7f5f2c23ecf4a,
title = "LAWC: Optimizing Write Cache Using Layout-Aware I/O Scheduling for All Flash Storage",
abstract = "Flash memory-based SSD-RAIDs are swiftly replacing conventional hard disk drives by exhibiting improved performance and stability, especially in I/O-intensive environments. However, the variations in latency and throughput occurring due to uncoordinated internal garbage collection cripples further boosting of performance. In addition, the unwanted variations in each SSD can influence the overall performance of the entire flash storage adversely. This performance bottleneck can be essentially reduced by an internal write cache in the RAID controller designed prudently by considering the crucial device characteristics. The state-of-the-art cache write for the RAID controller fails to incorporate device characteristics of flash memory-based SSDs and mitigates the performance gain. In this paper, we propose a novel cache design namely Layout-Aware Write Cache (LAWC) to overcome the performance barrier inculcated by independent garbage collections. LAWC implements (i) improved I/O scheduling for logically partitioned write caches, (ii) a destage write synchronization mechanism to allow individual write caches to flush write blocks into the SSD array in a coordinated manner, and (iii) a two-level hybrid cache algorithm utilizing small front level cache for the improved write cache efficiency. LAWC shows significant reduction in response time by 82.39 percent on RAID-0 and 68.51 percent on RAID-5 types of SSDs when compared with state-of-the-art write cache algorithms.",
keywords = "Flash memory, I/O scheduling, RAID, solid-state drive, storage system, write cache",
author = "Kalidas Ganesh and Youngjae Kim and Monobrata Debnath and Sungyong Park and Junghee Lee",
year = "2017",
month = "11",
day = "1",
doi = "10.1109/TC.2017.2707408",
language = "English",
volume = "66",
pages = "1890--1902",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "11",

}

TY - JOUR

T1 - LAWC

T2 - Optimizing Write Cache Using Layout-Aware I/O Scheduling for All Flash Storage

AU - Ganesh, Kalidas

AU - Kim, Youngjae

AU - Debnath, Monobrata

AU - Park, Sungyong

AU - Lee, Junghee

PY - 2017/11/1

Y1 - 2017/11/1

N2 - Flash memory-based SSD-RAIDs are swiftly replacing conventional hard disk drives by exhibiting improved performance and stability, especially in I/O-intensive environments. However, the variations in latency and throughput occurring due to uncoordinated internal garbage collection cripples further boosting of performance. In addition, the unwanted variations in each SSD can influence the overall performance of the entire flash storage adversely. This performance bottleneck can be essentially reduced by an internal write cache in the RAID controller designed prudently by considering the crucial device characteristics. The state-of-the-art cache write for the RAID controller fails to incorporate device characteristics of flash memory-based SSDs and mitigates the performance gain. In this paper, we propose a novel cache design namely Layout-Aware Write Cache (LAWC) to overcome the performance barrier inculcated by independent garbage collections. LAWC implements (i) improved I/O scheduling for logically partitioned write caches, (ii) a destage write synchronization mechanism to allow individual write caches to flush write blocks into the SSD array in a coordinated manner, and (iii) a two-level hybrid cache algorithm utilizing small front level cache for the improved write cache efficiency. LAWC shows significant reduction in response time by 82.39 percent on RAID-0 and 68.51 percent on RAID-5 types of SSDs when compared with state-of-the-art write cache algorithms.

AB - Flash memory-based SSD-RAIDs are swiftly replacing conventional hard disk drives by exhibiting improved performance and stability, especially in I/O-intensive environments. However, the variations in latency and throughput occurring due to uncoordinated internal garbage collection cripples further boosting of performance. In addition, the unwanted variations in each SSD can influence the overall performance of the entire flash storage adversely. This performance bottleneck can be essentially reduced by an internal write cache in the RAID controller designed prudently by considering the crucial device characteristics. The state-of-the-art cache write for the RAID controller fails to incorporate device characteristics of flash memory-based SSDs and mitigates the performance gain. In this paper, we propose a novel cache design namely Layout-Aware Write Cache (LAWC) to overcome the performance barrier inculcated by independent garbage collections. LAWC implements (i) improved I/O scheduling for logically partitioned write caches, (ii) a destage write synchronization mechanism to allow individual write caches to flush write blocks into the SSD array in a coordinated manner, and (iii) a two-level hybrid cache algorithm utilizing small front level cache for the improved write cache efficiency. LAWC shows significant reduction in response time by 82.39 percent on RAID-0 and 68.51 percent on RAID-5 types of SSDs when compared with state-of-the-art write cache algorithms.

KW - Flash memory

KW - I/O scheduling

KW - RAID

KW - solid-state drive

KW - storage system

KW - write cache

UR - http://www.scopus.com/inward/record.url?scp=85032383329&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85032383329&partnerID=8YFLogxK

U2 - 10.1109/TC.2017.2707408

DO - 10.1109/TC.2017.2707408

M3 - Article

AN - SCOPUS:85032383329

VL - 66

SP - 1890

EP - 1902

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 11

M1 - 7932926

ER -