You Can’t Hide You Can’t RunCitation formats

Standard

You Can’t Hide You Can’t Run : A Performance Assessment of Managed Applications on a NUMA Machine. / Papadakis, Orion; Zakkak, Foivos; Foutris, Nikolaos; Kotselidis, Christos-Efthymios.

2020. Paper presented at 17th International Conference on Managed Programming Languages and Runtimes, .

Research output: Contribution to conferencePaperpeer-review

Harvard

Papadakis, O, Zakkak, F, Foutris, N & Kotselidis, C-E 2020, 'You Can’t Hide You Can’t Run: A Performance Assessment of Managed Applications on a NUMA Machine', Paper presented at 17th International Conference on Managed Programming Languages and Runtimes, 4/11/20 - 6/11/20.

APA

Papadakis, O., Zakkak, F., Foutris, N., & Kotselidis, C-E. (Accepted/In press). You Can’t Hide You Can’t Run: A Performance Assessment of Managed Applications on a NUMA Machine. Paper presented at 17th International Conference on Managed Programming Languages and Runtimes, .

Vancouver

Papadakis O, Zakkak F, Foutris N, Kotselidis C-E. You Can’t Hide You Can’t Run: A Performance Assessment of Managed Applications on a NUMA Machine. 2020. Paper presented at 17th International Conference on Managed Programming Languages and Runtimes, .

Author

Papadakis, Orion ; Zakkak, Foivos ; Foutris, Nikolaos ; Kotselidis, Christos-Efthymios. / You Can’t Hide You Can’t Run : A Performance Assessment of Managed Applications on a NUMA Machine. Paper presented at 17th International Conference on Managed Programming Languages and Runtimes, .

Bibtex

@conference{d8808a8b91b440fa837685d83e13f88e,
title = "You Can{\textquoteright}t Hide You Can{\textquoteright}t Run: A Performance Assessment of Managed Applications on a NUMA Machine",
abstract = "The ever-growing demand for more memory capacity from applications has always been a challenging factor in computer architecture. The advent of the Non Unified Memory Access (NUMA) architecture has achieved to work around the physical constraints of a single processor by providing more system memory using pools of processors, each with their own memory elements, but with variable access times. However, the efficient exploitation of such computing systems is a non-trivial task for software engineers. We have observed that the performance of more than half of the applications picked from two distinct benchmark suites is negatively affected when running on a NUMA machine, in the absence of manual tuning. This finding motivated us to develop a new profiling tool, so called PerfUtil, to study, characterize and better understand why those benchmarks have sub-optimal performance on NUMA machines. PerfUtil's effectiveness is based on its ability to track numerous events throughout the system at the managed runtime system level, that, ultimately, assists in demystifying NUMA peculiarities and accurately characterize managed applications profiles",
keywords = "NUMA, JVM, Performance characterization",
author = "Orion Papadakis and Foivos Zakkak and Nikolaos Foutris and Christos-Efthymios Kotselidis",
year = "2020",
language = "English",
note = "17th International Conference on Managed Programming Languages and Runtimes, MPLR ; Conference date: 04-11-2020 Through 06-11-2020",
url = "https://mplr2020.cs.manchester.ac.uk/index.php",

}

RIS

TY - CONF

T1 - You Can’t Hide You Can’t Run

T2 - 17th International Conference on Managed Programming Languages and Runtimes

AU - Papadakis, Orion

AU - Zakkak, Foivos

AU - Foutris, Nikolaos

AU - Kotselidis, Christos-Efthymios

PY - 2020

Y1 - 2020

N2 - The ever-growing demand for more memory capacity from applications has always been a challenging factor in computer architecture. The advent of the Non Unified Memory Access (NUMA) architecture has achieved to work around the physical constraints of a single processor by providing more system memory using pools of processors, each with their own memory elements, but with variable access times. However, the efficient exploitation of such computing systems is a non-trivial task for software engineers. We have observed that the performance of more than half of the applications picked from two distinct benchmark suites is negatively affected when running on a NUMA machine, in the absence of manual tuning. This finding motivated us to develop a new profiling tool, so called PerfUtil, to study, characterize and better understand why those benchmarks have sub-optimal performance on NUMA machines. PerfUtil's effectiveness is based on its ability to track numerous events throughout the system at the managed runtime system level, that, ultimately, assists in demystifying NUMA peculiarities and accurately characterize managed applications profiles

AB - The ever-growing demand for more memory capacity from applications has always been a challenging factor in computer architecture. The advent of the Non Unified Memory Access (NUMA) architecture has achieved to work around the physical constraints of a single processor by providing more system memory using pools of processors, each with their own memory elements, but with variable access times. However, the efficient exploitation of such computing systems is a non-trivial task for software engineers. We have observed that the performance of more than half of the applications picked from two distinct benchmark suites is negatively affected when running on a NUMA machine, in the absence of manual tuning. This finding motivated us to develop a new profiling tool, so called PerfUtil, to study, characterize and better understand why those benchmarks have sub-optimal performance on NUMA machines. PerfUtil's effectiveness is based on its ability to track numerous events throughout the system at the managed runtime system level, that, ultimately, assists in demystifying NUMA peculiarities and accurately characterize managed applications profiles

KW - NUMA

KW - JVM

KW - Performance characterization

M3 - Paper

Y2 - 4 November 2020 through 6 November 2020

ER -