Performance evaluation and visualization of scientific applications using PMlib

Kazunori Mikami, Kenji Ono, Jorji Nonaka

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

Abstract

The computational performance of scientific applications on HPC systems is often much lower than user expectation based on the system's maximum performance specifications. To understand the basis for this performance gap, a multi-perspective evaluation is important. For instance, from the user perspective, correlating the theoretical computation coded as a source program with the actual computation workload produced by the compilers is valuable. From the system perspective, evaluating the characteristics of microarchitecture elements such as processor core and memory is of significance. An open source library called PMlib was developed to address these types of synthetic evaluations. PMlib provides an avenue for reporting the arithmetic/application workload explicitly coded in the source program, as well as the actually executed system workload. It also provides detailed utilization reports of processor-specific hardware including the categorized SIMD instruction statistics, the layered cache hit/miss rate, and the effective memory bandwidth, which are captured via hardware performance counters (HWPC). Using PMlib, users can conduct a synthetic analysis of application performance, and obtain useful feedback for further optimized execution of applications.

Original languageEnglish
Title of host publicationProceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages243-249
Number of pages7
ISBN (Electronic)9781538691847
DOIs
Publication statusPublished - Dec 26 2018
Event6th International Symposium on Computing and Networking Workshops, CANDARW 2018 - Takayama, Japan
Duration: Nov 27 2018Nov 30 2018

Publication series

NameProceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018

Conference

Conference6th International Symposium on Computing and Networking Workshops, CANDARW 2018
CountryJapan
CityTakayama
Period11/27/1811/30/18

Fingerprint

Performance Evaluation
Visualization
Workload
Data storage equipment
Hardware
Computer hardware
Evaluation
Hits
Open Source
Compiler
Cache
Statistics
Specifications
Feedback
Bandwidth
Performance evaluation
Specification

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Hardware and Architecture
  • Statistics, Probability and Uncertainty
  • Computer Science Applications

Cite this

Mikami, K., Ono, K., & Nonaka, J. (2018). Performance evaluation and visualization of scientific applications using PMlib. In Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018 (pp. 243-249). [8590907] (Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CANDARW.2018.00053

Performance evaluation and visualization of scientific applications using PMlib. / Mikami, Kazunori; Ono, Kenji; Nonaka, Jorji.

Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 243-249 8590907 (Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018).

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

Mikami, K, Ono, K & Nonaka, J 2018, Performance evaluation and visualization of scientific applications using PMlib. in Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018., 8590907, Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018, Institute of Electrical and Electronics Engineers Inc., pp. 243-249, 6th International Symposium on Computing and Networking Workshops, CANDARW 2018, Takayama, Japan, 11/27/18. https://doi.org/10.1109/CANDARW.2018.00053
Mikami K, Ono K, Nonaka J. Performance evaluation and visualization of scientific applications using PMlib. In Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 243-249. 8590907. (Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018). https://doi.org/10.1109/CANDARW.2018.00053
Mikami, Kazunori ; Ono, Kenji ; Nonaka, Jorji. / Performance evaluation and visualization of scientific applications using PMlib. Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 243-249 (Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018).
@inproceedings{0cf6ac182e2b4929a9235d63022c3051,
title = "Performance evaluation and visualization of scientific applications using PMlib",
abstract = "The computational performance of scientific applications on HPC systems is often much lower than user expectation based on the system's maximum performance specifications. To understand the basis for this performance gap, a multi-perspective evaluation is important. For instance, from the user perspective, correlating the theoretical computation coded as a source program with the actual computation workload produced by the compilers is valuable. From the system perspective, evaluating the characteristics of microarchitecture elements such as processor core and memory is of significance. An open source library called PMlib was developed to address these types of synthetic evaluations. PMlib provides an avenue for reporting the arithmetic/application workload explicitly coded in the source program, as well as the actually executed system workload. It also provides detailed utilization reports of processor-specific hardware including the categorized SIMD instruction statistics, the layered cache hit/miss rate, and the effective memory bandwidth, which are captured via hardware performance counters (HWPC). Using PMlib, users can conduct a synthetic analysis of application performance, and obtain useful feedback for further optimized execution of applications.",
author = "Kazunori Mikami and Kenji Ono and Jorji Nonaka",
year = "2018",
month = "12",
day = "26",
doi = "10.1109/CANDARW.2018.00053",
language = "English",
series = "Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "243--249",
booktitle = "Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018",
address = "United States",

}

TY - GEN

T1 - Performance evaluation and visualization of scientific applications using PMlib

AU - Mikami, Kazunori

AU - Ono, Kenji

AU - Nonaka, Jorji

PY - 2018/12/26

Y1 - 2018/12/26

N2 - The computational performance of scientific applications on HPC systems is often much lower than user expectation based on the system's maximum performance specifications. To understand the basis for this performance gap, a multi-perspective evaluation is important. For instance, from the user perspective, correlating the theoretical computation coded as a source program with the actual computation workload produced by the compilers is valuable. From the system perspective, evaluating the characteristics of microarchitecture elements such as processor core and memory is of significance. An open source library called PMlib was developed to address these types of synthetic evaluations. PMlib provides an avenue for reporting the arithmetic/application workload explicitly coded in the source program, as well as the actually executed system workload. It also provides detailed utilization reports of processor-specific hardware including the categorized SIMD instruction statistics, the layered cache hit/miss rate, and the effective memory bandwidth, which are captured via hardware performance counters (HWPC). Using PMlib, users can conduct a synthetic analysis of application performance, and obtain useful feedback for further optimized execution of applications.

AB - The computational performance of scientific applications on HPC systems is often much lower than user expectation based on the system's maximum performance specifications. To understand the basis for this performance gap, a multi-perspective evaluation is important. For instance, from the user perspective, correlating the theoretical computation coded as a source program with the actual computation workload produced by the compilers is valuable. From the system perspective, evaluating the characteristics of microarchitecture elements such as processor core and memory is of significance. An open source library called PMlib was developed to address these types of synthetic evaluations. PMlib provides an avenue for reporting the arithmetic/application workload explicitly coded in the source program, as well as the actually executed system workload. It also provides detailed utilization reports of processor-specific hardware including the categorized SIMD instruction statistics, the layered cache hit/miss rate, and the effective memory bandwidth, which are captured via hardware performance counters (HWPC). Using PMlib, users can conduct a synthetic analysis of application performance, and obtain useful feedback for further optimized execution of applications.

UR - http://www.scopus.com/inward/record.url?scp=85061450633&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061450633&partnerID=8YFLogxK

U2 - 10.1109/CANDARW.2018.00053

DO - 10.1109/CANDARW.2018.00053

M3 - Conference contribution

AN - SCOPUS:85061450633

T3 - Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018

SP - 243

EP - 249

BT - Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -