CApRI: CAche-conscious data reordering for irregular codes

Research output: Contribution to journalConference article

2 Citations (Scopus)

Abstract

Caches play a critical role in today's computer systems and optimizing their performance has been a critical objective in the last couple of decades. Unfortunately, compared to a plethora of work in software and hardware directed code/data optimizations, much less effort has been spent in understanding the fundamental characteristics of data access patterns exhibited by application programs and their interaction with the underlying cache hardware. Therefore, in general it is hard to reason about cache behavior of a program running on a target system. Motivated by this observation, we first set up a "locality model" that can help us determine the theoretical bounds of the cache misses caused by irregular data accesses. We then explain how this locality model can be used for different data locality optimization purposes. After that, based on our model, we propose a data reordering (data layout reorganization) scheme that can be applied after any existing data reordering schemes for irregular applications to improve cache performance by further reducing the cache misses. We evaluate the effectiveness of our scheme using a set of 8 programs with irregular data accesses, and show that it brings significant improvements over the state-of-the-art on two commercial multicore machines.

Original languageEnglish (US)
Pages (from-to)477-489
Number of pages13
JournalPerformance Evaluation Review
Volume42
Issue number1
DOIs
StatePublished - Jun 20 2014
EventACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems, SIGMETRICS 2014 - Austin, United States
Duration: Jun 16 2014Jun 20 2014

Fingerprint

Hardware
Application programs
Computer systems

All Science Journal Classification (ASJC) codes

  • Software
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

@article{bbcec0299ec645b7b3f9cb2b0daba479,
title = "CApRI: CAche-conscious data reordering for irregular codes",
abstract = "Caches play a critical role in today's computer systems and optimizing their performance has been a critical objective in the last couple of decades. Unfortunately, compared to a plethora of work in software and hardware directed code/data optimizations, much less effort has been spent in understanding the fundamental characteristics of data access patterns exhibited by application programs and their interaction with the underlying cache hardware. Therefore, in general it is hard to reason about cache behavior of a program running on a target system. Motivated by this observation, we first set up a {"}locality model{"} that can help us determine the theoretical bounds of the cache misses caused by irregular data accesses. We then explain how this locality model can be used for different data locality optimization purposes. After that, based on our model, we propose a data reordering (data layout reorganization) scheme that can be applied after any existing data reordering schemes for irregular applications to improve cache performance by further reducing the cache misses. We evaluate the effectiveness of our scheme using a set of 8 programs with irregular data accesses, and show that it brings significant improvements over the state-of-the-art on two commercial multicore machines.",
author = "Wei Ding and Kandemir, {Mahmut Taylan}",
year = "2014",
month = "6",
day = "20",
doi = "10.1145/2591971.2591992",
language = "English (US)",
volume = "42",
pages = "477--489",
journal = "Performance Evaluation Review",
issn = "0163-5999",
publisher = "Association for Computing Machinery (ACM)",
number = "1",

}

CApRI : CAche-conscious data reordering for irregular codes. / Ding, Wei; Kandemir, Mahmut Taylan.

In: Performance Evaluation Review, Vol. 42, No. 1, 20.06.2014, p. 477-489.

Research output: Contribution to journalConference article

TY - JOUR

T1 - CApRI

T2 - CAche-conscious data reordering for irregular codes

AU - Ding, Wei

AU - Kandemir, Mahmut Taylan

PY - 2014/6/20

Y1 - 2014/6/20

N2 - Caches play a critical role in today's computer systems and optimizing their performance has been a critical objective in the last couple of decades. Unfortunately, compared to a plethora of work in software and hardware directed code/data optimizations, much less effort has been spent in understanding the fundamental characteristics of data access patterns exhibited by application programs and their interaction with the underlying cache hardware. Therefore, in general it is hard to reason about cache behavior of a program running on a target system. Motivated by this observation, we first set up a "locality model" that can help us determine the theoretical bounds of the cache misses caused by irregular data accesses. We then explain how this locality model can be used for different data locality optimization purposes. After that, based on our model, we propose a data reordering (data layout reorganization) scheme that can be applied after any existing data reordering schemes for irregular applications to improve cache performance by further reducing the cache misses. We evaluate the effectiveness of our scheme using a set of 8 programs with irregular data accesses, and show that it brings significant improvements over the state-of-the-art on two commercial multicore machines.

AB - Caches play a critical role in today's computer systems and optimizing their performance has been a critical objective in the last couple of decades. Unfortunately, compared to a plethora of work in software and hardware directed code/data optimizations, much less effort has been spent in understanding the fundamental characteristics of data access patterns exhibited by application programs and their interaction with the underlying cache hardware. Therefore, in general it is hard to reason about cache behavior of a program running on a target system. Motivated by this observation, we first set up a "locality model" that can help us determine the theoretical bounds of the cache misses caused by irregular data accesses. We then explain how this locality model can be used for different data locality optimization purposes. After that, based on our model, we propose a data reordering (data layout reorganization) scheme that can be applied after any existing data reordering schemes for irregular applications to improve cache performance by further reducing the cache misses. We evaluate the effectiveness of our scheme using a set of 8 programs with irregular data accesses, and show that it brings significant improvements over the state-of-the-art on two commercial multicore machines.

UR - http://www.scopus.com/inward/record.url?scp=84955570053&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84955570053&partnerID=8YFLogxK

U2 - 10.1145/2591971.2591992

DO - 10.1145/2591971.2591992

M3 - Conference article

AN - SCOPUS:84955570053

VL - 42

SP - 477

EP - 489

JO - Performance Evaluation Review

JF - Performance Evaluation Review

SN - 0163-5999

IS - 1

ER -