Analytical memory bandwidth model for many-core processor based systems

Hyuk Jun Lee, Woo Cheol Cho, Eui Young Chung

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Many-core processor based systems gain popularity in high-performance parallel embedded applications. Estimating memory bandwidth requirement, i.e. external memory bandwidth, given various cache size for target parallel applications requires a prohibitively large simulation time. In this work, we propose an analytical model to quickly estimate the memory bandwidth for a given cache size and help exploring trade-offs between cache sizes and memory bandwidth requirement. We model the stochastic behavior of cache misses for a single cache as a random process. Using central limit theorems for identically or non-identically distributed random processes, we accurately estimate the collective cache misses from hundreds of processor cores and thus the total memory bandwidth requirement for the whole system. The results show that our model improves a speed of simulation time up to 200.4 times for 200 cores whereas its estimated results achieve less than 0.01% difference from the simulated ones for 200 cores in terms of accuracy.

Original languageEnglish
Pages (from-to)1461-1466
Number of pages6
Journalieice electronics express
Volume9
Issue number18
DOIs
Publication statusPublished - 2012 Nov 19

Fingerprint

central processing units
bandwidth
Bandwidth
Data storage equipment
random processes
Random processes
requirements
estimates
Analytical models
estimating
simulation
theorems

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Electrical and Electronic Engineering

Cite this

Lee, Hyuk Jun ; Cho, Woo Cheol ; Chung, Eui Young. / Analytical memory bandwidth model for many-core processor based systems. In: ieice electronics express. 2012 ; Vol. 9, No. 18. pp. 1461-1466.
@article{66df746ce8c248f2b24eb4ab1b9a669d,
title = "Analytical memory bandwidth model for many-core processor based systems",
abstract = "Many-core processor based systems gain popularity in high-performance parallel embedded applications. Estimating memory bandwidth requirement, i.e. external memory bandwidth, given various cache size for target parallel applications requires a prohibitively large simulation time. In this work, we propose an analytical model to quickly estimate the memory bandwidth for a given cache size and help exploring trade-offs between cache sizes and memory bandwidth requirement. We model the stochastic behavior of cache misses for a single cache as a random process. Using central limit theorems for identically or non-identically distributed random processes, we accurately estimate the collective cache misses from hundreds of processor cores and thus the total memory bandwidth requirement for the whole system. The results show that our model improves a speed of simulation time up to 200.4 times for 200 cores whereas its estimated results achieve less than 0.01{\%} difference from the simulated ones for 200 cores in terms of accuracy.",
author = "Lee, {Hyuk Jun} and Cho, {Woo Cheol} and Chung, {Eui Young}",
year = "2012",
month = "11",
day = "19",
doi = "10.1587/elex.9.1461",
language = "English",
volume = "9",
pages = "1461--1466",
journal = "IEICE Electronics Express",
issn = "1349-2543",
publisher = "The Institute of Electronics, Information and Communication Engineers (IEICE)",
number = "18",

}

Analytical memory bandwidth model for many-core processor based systems. / Lee, Hyuk Jun; Cho, Woo Cheol; Chung, Eui Young.

In: ieice electronics express, Vol. 9, No. 18, 19.11.2012, p. 1461-1466.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Analytical memory bandwidth model for many-core processor based systems

AU - Lee, Hyuk Jun

AU - Cho, Woo Cheol

AU - Chung, Eui Young

PY - 2012/11/19

Y1 - 2012/11/19

N2 - Many-core processor based systems gain popularity in high-performance parallel embedded applications. Estimating memory bandwidth requirement, i.e. external memory bandwidth, given various cache size for target parallel applications requires a prohibitively large simulation time. In this work, we propose an analytical model to quickly estimate the memory bandwidth for a given cache size and help exploring trade-offs between cache sizes and memory bandwidth requirement. We model the stochastic behavior of cache misses for a single cache as a random process. Using central limit theorems for identically or non-identically distributed random processes, we accurately estimate the collective cache misses from hundreds of processor cores and thus the total memory bandwidth requirement for the whole system. The results show that our model improves a speed of simulation time up to 200.4 times for 200 cores whereas its estimated results achieve less than 0.01% difference from the simulated ones for 200 cores in terms of accuracy.

AB - Many-core processor based systems gain popularity in high-performance parallel embedded applications. Estimating memory bandwidth requirement, i.e. external memory bandwidth, given various cache size for target parallel applications requires a prohibitively large simulation time. In this work, we propose an analytical model to quickly estimate the memory bandwidth for a given cache size and help exploring trade-offs between cache sizes and memory bandwidth requirement. We model the stochastic behavior of cache misses for a single cache as a random process. Using central limit theorems for identically or non-identically distributed random processes, we accurately estimate the collective cache misses from hundreds of processor cores and thus the total memory bandwidth requirement for the whole system. The results show that our model improves a speed of simulation time up to 200.4 times for 200 cores whereas its estimated results achieve less than 0.01% difference from the simulated ones for 200 cores in terms of accuracy.

UR - http://www.scopus.com/inward/record.url?scp=84869074704&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84869074704&partnerID=8YFLogxK

U2 - 10.1587/elex.9.1461

DO - 10.1587/elex.9.1461

M3 - Article

AN - SCOPUS:84869074704

VL - 9

SP - 1461

EP - 1466

JO - IEICE Electronics Express

JF - IEICE Electronics Express

SN - 1349-2543

IS - 18

ER -