On Improving the Write Responsiveness for Host-Aware SMR Drives

Ming Chang Yang; Yuan Hao Chang; Fenggang Wu; Tei Wei Kuo; David H Du

doi:10.1109/TC.2018.2845383

On Improving the Write Responsiveness for Host-Aware SMR Drives

Ming Chang Yang, Yuan Hao Chang, Fenggang Wu, Tei Wei Kuo, David H Du

Computer Science and Engineering

Research output: Contribution to journal › Article › peer-review

13 Scopus citations

Abstract

This paper presents a Virtual Persistent Cache design to remedy the long latency behavior and to ultimately improve the write responsiveness of the Host-Aware Shingled Magnetic Recording (HA-SMR) drives. Our design keeps the cost-effective model of the existing HA-SMR drives, but at the same time asks the great help from the host system for adaptively providing some computing and management resources to improve the drive performance when needed. The technical contribution is to trick the HA-SMR drives by smartly reshaping the access patterns to HA-SMR drives, so as to avoid the occurrences of long latencies in most cases and thus to ultimately improve the drive performance and responsiveness. We conduct experiments on real Seagate 8 TB HA-SMR drives to demonstrate the advantages of Virtual Persistent Cache over the real workloads from Microsoft Research Cambridge. The results show that the proposed design can remedy most of the long latencies and improve the drive performance by at least 58.11 percent, under the evaluated workloads.

Original language	English (US)
Article number	8387485
Pages (from-to)	111-124
Number of pages	14
Journal	IEEE Transactions on Computers
Volume	68
Issue number	1
DOIs	https://doi.org/10.1109/TC.2018.2845383
State	Published - Jan 1 2019

Bibliographical note

Funding Information:
This paper is an extended version of a paper appeared in International Conference on Computer-Aided Design (ICCAD), Irvine, California, USA, Nov. 13-16, 2017 [37]. This work is supported in part by Ministry of Science and Technology (MOST) under grant Nos. 105-2221-E-001-013-MY3, 105-2221-E-001-004-MY2, 105-3111-Y-001-041, and 104-2221-E-002-219-MY3, and is partially supported by US National Science Foundation (NSF) awards 130523, 1421913, 1439622 and 1525617.

Funding Information:
This paper is an extended version of a paper appeared in International Conference on Computer-Aided Design (ICCAD), Irvine, California, USA, Nov. 13-16, 2017 [37]. This work is supported in part by Ministry of Science and Technology (MOST) under grant Nos. 105-2221-E-001-013- MY3, 105-2221-E-001-004-MY2, 105-3111-Y-001-041, and 104-2221-E-002-219-MY3, and is partially supported by US National Science Foundation (NSF) awards 130523, 1421913, 1439622 and 1525617.

Publisher Copyright:
© 1968-2012 IEEE.

Keywords

Host-aware
long latency behavior
persistent cache
shingled magnetic recording (SMR)

Access

10.1109/TC.2018.2845383

OpenUrl availability

Full text

Cite this

@article{09d88cdd7c684e49abff79537ee05d5e,

title = "On Improving the Write Responsiveness for Host-Aware SMR Drives",

abstract = "This paper presents a Virtual Persistent Cache design to remedy the long latency behavior and to ultimately improve the write responsiveness of the Host-Aware Shingled Magnetic Recording (HA-SMR) drives. Our design keeps the cost-effective model of the existing HA-SMR drives, but at the same time asks the great help from the host system for adaptively providing some computing and management resources to improve the drive performance when needed. The technical contribution is to trick the HA-SMR drives by smartly reshaping the access patterns to HA-SMR drives, so as to avoid the occurrences of long latencies in most cases and thus to ultimately improve the drive performance and responsiveness. We conduct experiments on real Seagate 8 TB HA-SMR drives to demonstrate the advantages of Virtual Persistent Cache over the real workloads from Microsoft Research Cambridge. The results show that the proposed design can remedy most of the long latencies and improve the drive performance by at least 58.11 percent, under the evaluated workloads.",

keywords = "Host-aware, long latency behavior, persistent cache, shingled magnetic recording (SMR)",

author = "Yang, {Ming Chang} and Chang, {Yuan Hao} and Fenggang Wu and Kuo, {Tei Wei} and Du, {David H}",

note = "Funding Information: This paper is an extended version of a paper appeared in International Conference on Computer-Aided Design (ICCAD), Irvine, California, USA, Nov. 13-16, 2017 [37]. This work is supported in part by Ministry of Science and Technology (MOST) under grant Nos. 105-2221-E-001-013-MY3, 105-2221-E-001-004-MY2, 105-3111-Y-001-041, and 104-2221-E-002-219-MY3, and is partially supported by US National Science Foundation (NSF) awards 130523, 1421913, 1439622 and 1525617. Funding Information: This paper is an extended version of a paper appeared in International Conference on Computer-Aided Design (ICCAD), Irvine, California, USA, Nov. 13-16, 2017 [37]. This work is supported in part by Ministry of Science and Technology (MOST) under grant Nos. 105-2221-E-001-013- MY3, 105-2221-E-001-004-MY2, 105-3111-Y-001-041, and 104-2221-E-002-219-MY3, and is partially supported by US National Science Foundation (NSF) awards 130523, 1421913, 1439622 and 1525617. Publisher Copyright: {\textcopyright} 1968-2012 IEEE.",

year = "2019",

month = jan,

day = "1",

doi = "10.1109/TC.2018.2845383",

language = "English (US)",

volume = "68",

pages = "111--124",

journal = "IEEE Transactions on Computers",

issn = "0018-9340",

publisher = "IEEE Computer Society",

number = "1",

}

TY - JOUR

T1 - On Improving the Write Responsiveness for Host-Aware SMR Drives

AU - Yang, Ming Chang

AU - Chang, Yuan Hao

AU - Wu, Fenggang

AU - Kuo, Tei Wei

AU - Du, David H

N1 - Funding Information: This paper is an extended version of a paper appeared in International Conference on Computer-Aided Design (ICCAD), Irvine, California, USA, Nov. 13-16, 2017 [37]. This work is supported in part by Ministry of Science and Technology (MOST) under grant Nos. 105-2221-E-001-013-MY3, 105-2221-E-001-004-MY2, 105-3111-Y-001-041, and 104-2221-E-002-219-MY3, and is partially supported by US National Science Foundation (NSF) awards 130523, 1421913, 1439622 and 1525617. Funding Information: This paper is an extended version of a paper appeared in International Conference on Computer-Aided Design (ICCAD), Irvine, California, USA, Nov. 13-16, 2017 [37]. This work is supported in part by Ministry of Science and Technology (MOST) under grant Nos. 105-2221-E-001-013- MY3, 105-2221-E-001-004-MY2, 105-3111-Y-001-041, and 104-2221-E-002-219-MY3, and is partially supported by US National Science Foundation (NSF) awards 130523, 1421913, 1439622 and 1525617. Publisher Copyright: © 1968-2012 IEEE.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - This paper presents a Virtual Persistent Cache design to remedy the long latency behavior and to ultimately improve the write responsiveness of the Host-Aware Shingled Magnetic Recording (HA-SMR) drives. Our design keeps the cost-effective model of the existing HA-SMR drives, but at the same time asks the great help from the host system for adaptively providing some computing and management resources to improve the drive performance when needed. The technical contribution is to trick the HA-SMR drives by smartly reshaping the access patterns to HA-SMR drives, so as to avoid the occurrences of long latencies in most cases and thus to ultimately improve the drive performance and responsiveness. We conduct experiments on real Seagate 8 TB HA-SMR drives to demonstrate the advantages of Virtual Persistent Cache over the real workloads from Microsoft Research Cambridge. The results show that the proposed design can remedy most of the long latencies and improve the drive performance by at least 58.11 percent, under the evaluated workloads.

AB - This paper presents a Virtual Persistent Cache design to remedy the long latency behavior and to ultimately improve the write responsiveness of the Host-Aware Shingled Magnetic Recording (HA-SMR) drives. Our design keeps the cost-effective model of the existing HA-SMR drives, but at the same time asks the great help from the host system for adaptively providing some computing and management resources to improve the drive performance when needed. The technical contribution is to trick the HA-SMR drives by smartly reshaping the access patterns to HA-SMR drives, so as to avoid the occurrences of long latencies in most cases and thus to ultimately improve the drive performance and responsiveness. We conduct experiments on real Seagate 8 TB HA-SMR drives to demonstrate the advantages of Virtual Persistent Cache over the real workloads from Microsoft Research Cambridge. The results show that the proposed design can remedy most of the long latencies and improve the drive performance by at least 58.11 percent, under the evaluated workloads.

KW - Host-aware

KW - long latency behavior

KW - persistent cache

KW - shingled magnetic recording (SMR)

UR - http://www.scopus.com/inward/record.url?scp=85048599033&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048599033&partnerID=8YFLogxK

U2 - 10.1109/TC.2018.2845383

DO - 10.1109/TC.2018.2845383

M3 - Article

AN - SCOPUS:85048599033

SN - 0018-9340

VL - 68

SP - 111

EP - 124

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

IS - 1

M1 - 8387485

ER -

On Improving the Write Responsiveness for Host-Aware SMR Drives

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this