Architecture-aware automatic computation offload for native applications

Gwangmu Lee, Hyunjoon Park, Seonyeong Heo, Kyung Ah Chang, Hyogun Lee, Hanjun Kim

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

15 Citations (Scopus)

Abstract

Although mobile devices have been evolved enough to support complex mobile programs, performance of the mobile devices is lagging behind performance of servers. To bridge the performance gap, computation offloading allows a mobile device to remotely execute heavy tasks at servers. However, due to architectural differences between mobile devices and servers, most existing computation offloading systems rely on virtual machines, so they cannot offload native applications. Some offloading systems can offload native mobile applications, but their applicability is limited to well-analyzable simple applications. This work presents automatic cross-architecture computation offloading for general-purpose native applications with a prototype framework that is called Native Offloader. At compile-time, Native Offloader automatically finds heavy tasks without any annotation, and generates offloading-enabled native binaries with memory unification for a mobile device and a server. At run-time, Native Offloader efficiently supports seamless migration between the mobile device and the server with a unified virtual address space and communication optimization. Native Offloader automatically offloads 17 native C applications from SPEC CPU2000 and CPU2006 benchmark suites without a virtual machine, and achieves a geomean program speedup of 6.42× and battery saving of 82.0%.

Original languageEnglish
Title of host publicationProceedings - 48th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2015
PublisherIEEE Computer Society
Pages521-532
Number of pages12
ISBN (Electronic)9781450340342
DOIs
Publication statusPublished - 2015 Dec 5
Event48th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2015 - Waikiki, United States
Duration: 2015 Dec 52015 Dec 9

Publication series

NameProceedings of the Annual International Symposium on Microarchitecture, MICRO
Volume05-09-December-2015
ISSN (Print)1072-4451

Other

Other48th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2015
CountryUnited States
CityWaikiki
Period15/12/515/12/9

Fingerprint

Mobile devices
Servers
Virtual addresses
Computer systems
Data storage equipment
Communication
Virtual machine

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture

Cite this

Lee, G., Park, H., Heo, S., Chang, K. A., Lee, H., & Kim, H. (2015). Architecture-aware automatic computation offload for native applications. In Proceedings - 48th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2015 (pp. 521-532). (Proceedings of the Annual International Symposium on Microarchitecture, MICRO; Vol. 05-09-December-2015). IEEE Computer Society. https://doi.org/10.1145/2830772.2830833
Lee, Gwangmu ; Park, Hyunjoon ; Heo, Seonyeong ; Chang, Kyung Ah ; Lee, Hyogun ; Kim, Hanjun. / Architecture-aware automatic computation offload for native applications. Proceedings - 48th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2015. IEEE Computer Society, 2015. pp. 521-532 (Proceedings of the Annual International Symposium on Microarchitecture, MICRO).
@inproceedings{3f72866ef1404a7584911f34c7b7b58f,
title = "Architecture-aware automatic computation offload for native applications",
abstract = "Although mobile devices have been evolved enough to support complex mobile programs, performance of the mobile devices is lagging behind performance of servers. To bridge the performance gap, computation offloading allows a mobile device to remotely execute heavy tasks at servers. However, due to architectural differences between mobile devices and servers, most existing computation offloading systems rely on virtual machines, so they cannot offload native applications. Some offloading systems can offload native mobile applications, but their applicability is limited to well-analyzable simple applications. This work presents automatic cross-architecture computation offloading for general-purpose native applications with a prototype framework that is called Native Offloader. At compile-time, Native Offloader automatically finds heavy tasks without any annotation, and generates offloading-enabled native binaries with memory unification for a mobile device and a server. At run-time, Native Offloader efficiently supports seamless migration between the mobile device and the server with a unified virtual address space and communication optimization. Native Offloader automatically offloads 17 native C applications from SPEC CPU2000 and CPU2006 benchmark suites without a virtual machine, and achieves a geomean program speedup of 6.42× and battery saving of 82.0{\%}.",
author = "Gwangmu Lee and Hyunjoon Park and Seonyeong Heo and Chang, {Kyung Ah} and Hyogun Lee and Hanjun Kim",
year = "2015",
month = "12",
day = "5",
doi = "10.1145/2830772.2830833",
language = "English",
series = "Proceedings of the Annual International Symposium on Microarchitecture, MICRO",
publisher = "IEEE Computer Society",
pages = "521--532",
booktitle = "Proceedings - 48th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2015",
address = "United States",

}

Lee, G, Park, H, Heo, S, Chang, KA, Lee, H & Kim, H 2015, Architecture-aware automatic computation offload for native applications. in Proceedings - 48th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2015. Proceedings of the Annual International Symposium on Microarchitecture, MICRO, vol. 05-09-December-2015, IEEE Computer Society, pp. 521-532, 48th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2015, Waikiki, United States, 15/12/5. https://doi.org/10.1145/2830772.2830833

Architecture-aware automatic computation offload for native applications. / Lee, Gwangmu; Park, Hyunjoon; Heo, Seonyeong; Chang, Kyung Ah; Lee, Hyogun; Kim, Hanjun.

Proceedings - 48th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2015. IEEE Computer Society, 2015. p. 521-532 (Proceedings of the Annual International Symposium on Microarchitecture, MICRO; Vol. 05-09-December-2015).

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

TY - GEN

T1 - Architecture-aware automatic computation offload for native applications

AU - Lee, Gwangmu

AU - Park, Hyunjoon

AU - Heo, Seonyeong

AU - Chang, Kyung Ah

AU - Lee, Hyogun

AU - Kim, Hanjun

PY - 2015/12/5

Y1 - 2015/12/5

N2 - Although mobile devices have been evolved enough to support complex mobile programs, performance of the mobile devices is lagging behind performance of servers. To bridge the performance gap, computation offloading allows a mobile device to remotely execute heavy tasks at servers. However, due to architectural differences between mobile devices and servers, most existing computation offloading systems rely on virtual machines, so they cannot offload native applications. Some offloading systems can offload native mobile applications, but their applicability is limited to well-analyzable simple applications. This work presents automatic cross-architecture computation offloading for general-purpose native applications with a prototype framework that is called Native Offloader. At compile-time, Native Offloader automatically finds heavy tasks without any annotation, and generates offloading-enabled native binaries with memory unification for a mobile device and a server. At run-time, Native Offloader efficiently supports seamless migration between the mobile device and the server with a unified virtual address space and communication optimization. Native Offloader automatically offloads 17 native C applications from SPEC CPU2000 and CPU2006 benchmark suites without a virtual machine, and achieves a geomean program speedup of 6.42× and battery saving of 82.0%.

AB - Although mobile devices have been evolved enough to support complex mobile programs, performance of the mobile devices is lagging behind performance of servers. To bridge the performance gap, computation offloading allows a mobile device to remotely execute heavy tasks at servers. However, due to architectural differences between mobile devices and servers, most existing computation offloading systems rely on virtual machines, so they cannot offload native applications. Some offloading systems can offload native mobile applications, but their applicability is limited to well-analyzable simple applications. This work presents automatic cross-architecture computation offloading for general-purpose native applications with a prototype framework that is called Native Offloader. At compile-time, Native Offloader automatically finds heavy tasks without any annotation, and generates offloading-enabled native binaries with memory unification for a mobile device and a server. At run-time, Native Offloader efficiently supports seamless migration between the mobile device and the server with a unified virtual address space and communication optimization. Native Offloader automatically offloads 17 native C applications from SPEC CPU2000 and CPU2006 benchmark suites without a virtual machine, and achieves a geomean program speedup of 6.42× and battery saving of 82.0%.

UR - http://www.scopus.com/inward/record.url?scp=84959916960&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959916960&partnerID=8YFLogxK

U2 - 10.1145/2830772.2830833

DO - 10.1145/2830772.2830833

M3 - Conference contribution

AN - SCOPUS:84959916960

T3 - Proceedings of the Annual International Symposium on Microarchitecture, MICRO

SP - 521

EP - 532

BT - Proceedings - 48th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2015

PB - IEEE Computer Society

ER -

Lee G, Park H, Heo S, Chang KA, Lee H, Kim H. Architecture-aware automatic computation offload for native applications. In Proceedings - 48th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2015. IEEE Computer Society. 2015. p. 521-532. (Proceedings of the Annual International Symposium on Microarchitecture, MICRO). https://doi.org/10.1145/2830772.2830833