A library-based tool to improve CPU assignment for multicore processor-based pervasive servers

Yuki Ohno, Sayaka Akioka, Midori Sugaya, Tatsuo Nakajima

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

Abstract

This paper proposes SPLiT (Scalable Performance Library Tool) as the methodology to improve performance of applications on multicore processors, through CPU, and cache optimizations. SPLiT analyzes applications based on cycle counts, and cache misses, and predicts behavior of the applications according to the analysis. With these acquired knowledge on the target application, SPLiT improves CPU resource allocation determined by operating systems. SPLiT, and SPLiT library are designed to achieve resource optimizations based on the combination of hardware information collected by Operating System kernel, and software information collected by applications. Additionally, all the features of SPLiT are immediately available with the small modifications to the application source codes, and the modifications are basically to implant SPLiT library calls into the source code. This simple requirement diminishes programmers' difficulties for performance tuning of multicore applications. Through empirical experiments, we validated the efficiency of SPLiT, and observed that performance of the web application was improved by 26%.

Original languageEnglish
Title of host publicationProceedings - 16th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2010
Pages114-123
Number of pages10
DOIs
Publication statusPublished - 2010
Externally publishedYes
Event16th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2010 - Macau
Duration: 2010 Aug 232010 Aug 25

Other

Other16th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2010
CityMacau
Period10/8/2310/8/25

Fingerprint

Program processors
Servers
Computer operating systems
Resource allocation
Tuning
Hardware

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Computer Science Applications
  • Hardware and Architecture

Cite this

Ohno, Y., Akioka, S., Sugaya, M., & Nakajima, T. (2010). A library-based tool to improve CPU assignment for multicore processor-based pervasive servers. In Proceedings - 16th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2010 (pp. 114-123). [5591305] https://doi.org/10.1109/RTCSA.2010.21

A library-based tool to improve CPU assignment for multicore processor-based pervasive servers. / Ohno, Yuki; Akioka, Sayaka; Sugaya, Midori; Nakajima, Tatsuo.

Proceedings - 16th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2010. 2010. p. 114-123 5591305.

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

Ohno, Y, Akioka, S, Sugaya, M & Nakajima, T 2010, A library-based tool to improve CPU assignment for multicore processor-based pervasive servers. in Proceedings - 16th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2010., 5591305, pp. 114-123, 16th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2010, Macau, 10/8/23. https://doi.org/10.1109/RTCSA.2010.21
Ohno Y, Akioka S, Sugaya M, Nakajima T. A library-based tool to improve CPU assignment for multicore processor-based pervasive servers. In Proceedings - 16th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2010. 2010. p. 114-123. 5591305 https://doi.org/10.1109/RTCSA.2010.21
Ohno, Yuki ; Akioka, Sayaka ; Sugaya, Midori ; Nakajima, Tatsuo. / A library-based tool to improve CPU assignment for multicore processor-based pervasive servers. Proceedings - 16th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2010. 2010. pp. 114-123
@inproceedings{710340b046594840be0f2f5850764ad8,
title = "A library-based tool to improve CPU assignment for multicore processor-based pervasive servers",
abstract = "This paper proposes SPLiT (Scalable Performance Library Tool) as the methodology to improve performance of applications on multicore processors, through CPU, and cache optimizations. SPLiT analyzes applications based on cycle counts, and cache misses, and predicts behavior of the applications according to the analysis. With these acquired knowledge on the target application, SPLiT improves CPU resource allocation determined by operating systems. SPLiT, and SPLiT library are designed to achieve resource optimizations based on the combination of hardware information collected by Operating System kernel, and software information collected by applications. Additionally, all the features of SPLiT are immediately available with the small modifications to the application source codes, and the modifications are basically to implant SPLiT library calls into the source code. This simple requirement diminishes programmers' difficulties for performance tuning of multicore applications. Through empirical experiments, we validated the efficiency of SPLiT, and observed that performance of the web application was improved by 26{\%}.",
author = "Yuki Ohno and Sayaka Akioka and Midori Sugaya and Tatsuo Nakajima",
year = "2010",
doi = "10.1109/RTCSA.2010.21",
language = "English",
isbn = "9780769541556",
pages = "114--123",
booktitle = "Proceedings - 16th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2010",

}

TY - GEN

T1 - A library-based tool to improve CPU assignment for multicore processor-based pervasive servers

AU - Ohno, Yuki

AU - Akioka, Sayaka

AU - Sugaya, Midori

AU - Nakajima, Tatsuo

PY - 2010

Y1 - 2010

N2 - This paper proposes SPLiT (Scalable Performance Library Tool) as the methodology to improve performance of applications on multicore processors, through CPU, and cache optimizations. SPLiT analyzes applications based on cycle counts, and cache misses, and predicts behavior of the applications according to the analysis. With these acquired knowledge on the target application, SPLiT improves CPU resource allocation determined by operating systems. SPLiT, and SPLiT library are designed to achieve resource optimizations based on the combination of hardware information collected by Operating System kernel, and software information collected by applications. Additionally, all the features of SPLiT are immediately available with the small modifications to the application source codes, and the modifications are basically to implant SPLiT library calls into the source code. This simple requirement diminishes programmers' difficulties for performance tuning of multicore applications. Through empirical experiments, we validated the efficiency of SPLiT, and observed that performance of the web application was improved by 26%.

AB - This paper proposes SPLiT (Scalable Performance Library Tool) as the methodology to improve performance of applications on multicore processors, through CPU, and cache optimizations. SPLiT analyzes applications based on cycle counts, and cache misses, and predicts behavior of the applications according to the analysis. With these acquired knowledge on the target application, SPLiT improves CPU resource allocation determined by operating systems. SPLiT, and SPLiT library are designed to achieve resource optimizations based on the combination of hardware information collected by Operating System kernel, and software information collected by applications. Additionally, all the features of SPLiT are immediately available with the small modifications to the application source codes, and the modifications are basically to implant SPLiT library calls into the source code. This simple requirement diminishes programmers' difficulties for performance tuning of multicore applications. Through empirical experiments, we validated the efficiency of SPLiT, and observed that performance of the web application was improved by 26%.

UR - http://www.scopus.com/inward/record.url?scp=78449311065&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78449311065&partnerID=8YFLogxK

U2 - 10.1109/RTCSA.2010.21

DO - 10.1109/RTCSA.2010.21

M3 - Conference contribution

AN - SCOPUS:78449311065

SN - 9780769541556

SP - 114

EP - 123

BT - Proceedings - 16th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2010

ER -