Accurately evaluating application performance in simulated hybrid multi-tasking systems

Kyle Rupnow; Jacob Adriaens; Wenyin Fu; Katherine Compton

doi:10.1145/1723112.1723136

Accurately evaluating application performance in simulated hybrid multi-tasking systems

Kyle Rupnow^*, Jacob Adriaens, Wenyin Fu, Katherine Compton

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Evaluating the performance of reconfigurable computing applications in multi-tasking systems using simulation (as can be needed in early design-space exploration) faces several challenges. The complexity of full-system, cycle-accurate simulation prevents executing applications of any appreciable size to completion. One must sample only a portion of execution; yet unless care is taken, the measured performance for the sampled interval will not be indicative of the complete execution. Although this is generally a problem for simulation-based evaluation, the problem is exacerbated for multi-tasking systems. This paper therefore presents work to develop a performance evaluation methodology that accurately measures hybrid (both hardware and software) application performance, accounts for additional overhead introduced by hybrid resource management (such as run-time allocation of reconfigurable hardware), and correctly compensates for momentary imbalances in processor time allocation that are only artifacts of the (necessarily) short simulated execution timespan and would balance out over time.

Original language	English (US)
Title of host publication	FPGA'10 - Proceedings of the 18th ACM SIGDA International Symposium on Field-Programmable Gate Arrays
Pages	135-143
Number of pages	9
DOIs	https://doi.org/10.1145/1723112.1723136
State	Published - 2010
Externally published	Yes
Event	18th ACM SIGDA International Symposium on Field-Programmable Gate Arrays, FPGA'10 - Monterey, CA, United States Duration: Feb 21 2010 → Feb 23 2010

Publication series

Name	ACM/SIGDA International Symposium on Field Programmable Gate Arrays - FPGA

Conference

Conference	18th ACM SIGDA International Symposium on Field-Programmable Gate Arrays, FPGA'10
Country/Territory	United States
City	Monterey, CA
Period	2/21/10 → 2/23/10

Keywords

Full system simulation
Heterogeneous systems
Hybrid systems
Multi-tasking systems
Performance evaluation

ASJC Scopus subject areas

Computer Science(all)

Access to Document

10.1145/1723112.1723136

Cite this

Rupnow, K., Adriaens, J., Fu, W., & Compton, K. (2010). Accurately evaluating application performance in simulated hybrid multi-tasking systems. In FPGA'10 - Proceedings of the 18th ACM SIGDA International Symposium on Field-Programmable Gate Arrays (pp. 135-143). (ACM/SIGDA International Symposium on Field Programmable Gate Arrays - FPGA). https://doi.org/10.1145/1723112.1723136

@inproceedings{69f7cbd595c14449b19f1a0749bd8736,

title = "Accurately evaluating application performance in simulated hybrid multi-tasking systems",

abstract = "Evaluating the performance of reconfigurable computing applications in multi-tasking systems using simulation (as can be needed in early design-space exploration) faces several challenges. The complexity of full-system, cycle-accurate simulation prevents executing applications of any appreciable size to completion. One must sample only a portion of execution; yet unless care is taken, the measured performance for the sampled interval will not be indicative of the complete execution. Although this is generally a problem for simulation-based evaluation, the problem is exacerbated for multi-tasking systems. This paper therefore presents work to develop a performance evaluation methodology that accurately measures hybrid (both hardware and software) application performance, accounts for additional overhead introduced by hybrid resource management (such as run-time allocation of reconfigurable hardware), and correctly compensates for momentary imbalances in processor time allocation that are only artifacts of the (necessarily) short simulated execution timespan and would balance out over time.",

keywords = "Full system simulation, Heterogeneous systems, Hybrid systems, Multi-tasking systems, Performance evaluation",

author = "Kyle Rupnow and Jacob Adriaens and Wenyin Fu and Katherine Compton",

note = "Copyright: Copyright 2010 Elsevier B.V., All rights reserved.; 18th ACM SIGDA International Symposium on Field-Programmable Gate Arrays, FPGA'10 ; Conference date: 21-02-2010 Through 23-02-2010",

year = "2010",

doi = "10.1145/1723112.1723136",

language = "English (US)",

isbn = "9781605589114",

series = "ACM/SIGDA International Symposium on Field Programmable Gate Arrays - FPGA",

pages = "135--143",

booktitle = "FPGA'10 - Proceedings of the 18th ACM SIGDA International Symposium on Field-Programmable Gate Arrays",

}

Rupnow, K, Adriaens, J, Fu, W & Compton, K 2010, Accurately evaluating application performance in simulated hybrid multi-tasking systems. in FPGA'10 - Proceedings of the 18th ACM SIGDA International Symposium on Field-Programmable Gate Arrays. ACM/SIGDA International Symposium on Field Programmable Gate Arrays - FPGA, pp. 135-143, 18th ACM SIGDA International Symposium on Field-Programmable Gate Arrays, FPGA'10, Monterey, CA, United States, 2/21/10. https://doi.org/10.1145/1723112.1723136

Accurately evaluating application performance in simulated hybrid multi-tasking systems. / Rupnow, Kyle; Adriaens, Jacob; Fu, Wenyin et al.
FPGA'10 - Proceedings of the 18th ACM SIGDA International Symposium on Field-Programmable Gate Arrays. 2010. p. 135-143 (ACM/SIGDA International Symposium on Field Programmable Gate Arrays - FPGA).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Accurately evaluating application performance in simulated hybrid multi-tasking systems

AU - Rupnow, Kyle

AU - Adriaens, Jacob

AU - Fu, Wenyin

AU - Compton, Katherine

PY - 2010

Y1 - 2010

N2 - Evaluating the performance of reconfigurable computing applications in multi-tasking systems using simulation (as can be needed in early design-space exploration) faces several challenges. The complexity of full-system, cycle-accurate simulation prevents executing applications of any appreciable size to completion. One must sample only a portion of execution; yet unless care is taken, the measured performance for the sampled interval will not be indicative of the complete execution. Although this is generally a problem for simulation-based evaluation, the problem is exacerbated for multi-tasking systems. This paper therefore presents work to develop a performance evaluation methodology that accurately measures hybrid (both hardware and software) application performance, accounts for additional overhead introduced by hybrid resource management (such as run-time allocation of reconfigurable hardware), and correctly compensates for momentary imbalances in processor time allocation that are only artifacts of the (necessarily) short simulated execution timespan and would balance out over time.

AB - Evaluating the performance of reconfigurable computing applications in multi-tasking systems using simulation (as can be needed in early design-space exploration) faces several challenges. The complexity of full-system, cycle-accurate simulation prevents executing applications of any appreciable size to completion. One must sample only a portion of execution; yet unless care is taken, the measured performance for the sampled interval will not be indicative of the complete execution. Although this is generally a problem for simulation-based evaluation, the problem is exacerbated for multi-tasking systems. This paper therefore presents work to develop a performance evaluation methodology that accurately measures hybrid (both hardware and software) application performance, accounts for additional overhead introduced by hybrid resource management (such as run-time allocation of reconfigurable hardware), and correctly compensates for momentary imbalances in processor time allocation that are only artifacts of the (necessarily) short simulated execution timespan and would balance out over time.

KW - Full system simulation

KW - Heterogeneous systems

KW - Hybrid systems

KW - Multi-tasking systems

KW - Performance evaluation

UR - http://www.scopus.com/inward/record.url?scp=77951517336&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951517336&partnerID=8YFLogxK

U2 - 10.1145/1723112.1723136

DO - 10.1145/1723112.1723136

M3 - Conference contribution

AN - SCOPUS:77951517336

SN - 9781605589114

T3 - ACM/SIGDA International Symposium on Field Programmable Gate Arrays - FPGA

SP - 135

EP - 143

BT - FPGA'10 - Proceedings of the 18th ACM SIGDA International Symposium on Field-Programmable Gate Arrays

T2 - 18th ACM SIGDA International Symposium on Field-Programmable Gate Arrays, FPGA'10

Y2 - 21 February 2010 through 23 February 2010

ER -

Accurately evaluating application performance in simulated hybrid multi-tasking systems

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this