Framework for efficient and flexible scheduling of flash memory operations

Bryan S. Kim, Yonggun Lee, Sang Lyul Min

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Flash memory-based storages are used in a wide range of systems from small mobile devices to large-scale system servers. The performance demand from applications and the technology of flash memory vary widely from one system to another, making it difficult to design a universal flash memory scheduler for all systems. In this paper, we present a framework for efficient and flexible flash memory scheduling and compare a software scheduler based on the framework against an optimized hardware scheduler. The throughput of the software scheduler achieves more than 97% of that of the hardware scheduler across workloads we consider. We further highlight the extensibility of the framework with a case study on fair queueing scheduling.

Original languageEnglish (US)
Title of host publicationNVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538617687
DOIs
StatePublished - Oct 10 2017
Externally publishedYes
Event6th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2017 - Hsinchu, Taiwan, Province of China
Duration: Aug 16 2017Aug 18 2017

Publication series

NameNVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium

Conference

Conference6th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2017
CountryTaiwan, Province of China
CityHsinchu
Period8/16/178/18/17

Fingerprint

Flash memory
Scheduling
Hardware
Mobile devices
Large scale systems
Servers
Throughput

Keywords

  • Flash Memory
  • Flash Memory Scheduling
  • Framework for Flexibility

ASJC Scopus subject areas

  • Hardware and Architecture

Cite this

Kim, B. S., Lee, Y., & Min, S. L. (2017). Framework for efficient and flexible scheduling of flash memory operations. In NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium [8064480] (NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/NVMSA.2017.8064480

Framework for efficient and flexible scheduling of flash memory operations. / Kim, Bryan S.; Lee, Yonggun; Min, Sang Lyul.

NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium. Institute of Electrical and Electronics Engineers Inc., 2017. 8064480 (NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kim, BS, Lee, Y & Min, SL 2017, Framework for efficient and flexible scheduling of flash memory operations. in NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium., 8064480, NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium, Institute of Electrical and Electronics Engineers Inc., 6th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2017, Hsinchu, Taiwan, Province of China, 8/16/17. https://doi.org/10.1109/NVMSA.2017.8064480
Kim BS, Lee Y, Min SL. Framework for efficient and flexible scheduling of flash memory operations. In NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium. Institute of Electrical and Electronics Engineers Inc. 2017. 8064480. (NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium). https://doi.org/10.1109/NVMSA.2017.8064480
Kim, Bryan S. ; Lee, Yonggun ; Min, Sang Lyul. / Framework for efficient and flexible scheduling of flash memory operations. NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium. Institute of Electrical and Electronics Engineers Inc., 2017. (NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium).
@inproceedings{b42259dee9594121bde149e446175028,
title = "Framework for efficient and flexible scheduling of flash memory operations",
abstract = "Flash memory-based storages are used in a wide range of systems from small mobile devices to large-scale system servers. The performance demand from applications and the technology of flash memory vary widely from one system to another, making it difficult to design a universal flash memory scheduler for all systems. In this paper, we present a framework for efficient and flexible flash memory scheduling and compare a software scheduler based on the framework against an optimized hardware scheduler. The throughput of the software scheduler achieves more than 97{\%} of that of the hardware scheduler across workloads we consider. We further highlight the extensibility of the framework with a case study on fair queueing scheduling.",
keywords = "Flash Memory, Flash Memory Scheduling, Framework for Flexibility",
author = "Kim, {Bryan S.} and Yonggun Lee and Min, {Sang Lyul}",
year = "2017",
month = "10",
day = "10",
doi = "10.1109/NVMSA.2017.8064480",
language = "English (US)",
series = "NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium",

}

TY - GEN

T1 - Framework for efficient and flexible scheduling of flash memory operations

AU - Kim, Bryan S.

AU - Lee, Yonggun

AU - Min, Sang Lyul

PY - 2017/10/10

Y1 - 2017/10/10

N2 - Flash memory-based storages are used in a wide range of systems from small mobile devices to large-scale system servers. The performance demand from applications and the technology of flash memory vary widely from one system to another, making it difficult to design a universal flash memory scheduler for all systems. In this paper, we present a framework for efficient and flexible flash memory scheduling and compare a software scheduler based on the framework against an optimized hardware scheduler. The throughput of the software scheduler achieves more than 97% of that of the hardware scheduler across workloads we consider. We further highlight the extensibility of the framework with a case study on fair queueing scheduling.

AB - Flash memory-based storages are used in a wide range of systems from small mobile devices to large-scale system servers. The performance demand from applications and the technology of flash memory vary widely from one system to another, making it difficult to design a universal flash memory scheduler for all systems. In this paper, we present a framework for efficient and flexible flash memory scheduling and compare a software scheduler based on the framework against an optimized hardware scheduler. The throughput of the software scheduler achieves more than 97% of that of the hardware scheduler across workloads we consider. We further highlight the extensibility of the framework with a case study on fair queueing scheduling.

KW - Flash Memory

KW - Flash Memory Scheduling

KW - Framework for Flexibility

UR - http://www.scopus.com/inward/record.url?scp=85034744468&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85034744468&partnerID=8YFLogxK

U2 - 10.1109/NVMSA.2017.8064480

DO - 10.1109/NVMSA.2017.8064480

M3 - Conference contribution

AN - SCOPUS:85034744468

T3 - NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium

BT - NVMSA 2017 - 6th IEEE Non-Volatile Memory Systems and Applications Symposium

PB - Institute of Electrical and Electronics Engineers Inc.

ER -