The performance of runtime data cache prefetching in a dynamic optimization system

Jiwei Lu; H. Chen; Rao Fu; Wei Chung Hsu; B. Othmer; Pen Chung Yew; Dong Yuan Chen

doi:10.1109/MICRO.2003.1253194

The performance of runtime data cache prefetching in a dynamic optimization system

Jiwei Lu, H. Chen, Rao Fu, Wei Chung Hsu, B. Othmer, Pen Chung Yew, Dong Yuan Chen

Computer Science and Engineering

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

62 Scopus citations

Abstract

Traditional software controlled data cache prefetching is often ineffective due to the lack of runtime cache miss and miss address information. To overcome this limitation, we implement runtime data cache prefetching in the dynamic optimization system ADORE (ADaptive Object code Reoptimization). Its performance has been compared with static software prefetching on the SPEC2000 benchmark suite. Runtime cache prefetching shows better performance. On an Itanium 2 based Linux workstation, it can increase performance by more than 20% over static prefetching on some benchmarks. For benchmarks that do not benefit from prefetching, the runtime optimization system adds only 1%-2% overhead. We have also collected cache miss profiles to guide static data cache prefetching in the ORC compiler. With that information the compiler can effectively avoid generating prefetches for loops that hit well in the data cache.

Original language	English (US)
Title of host publication	Proceedings - 36th International Symposium on Microarchitecture, MICRO 2003
Publisher	IEEE Computer Society
Pages	180-190
Number of pages	11
ISBN (Electronic)	076952043X
DOIs	https://doi.org/10.1109/MICRO.2003.1253194
State	Published - 2003
Event	36th International Symposium on Microarchitecture, MICRO 2003 - San Diego, United States Duration: Dec 3 2003 → Dec 5 2003

Publication series

Name	Proceedings of the Annual International Symposium on Microarchitecture, MICRO
Volume	2003-January
ISSN (Print)	1072-4451

Other

Other	36th International Symposium on Microarchitecture, MICRO 2003
Country/Territory	United States
City	San Diego
Period	12/3/03 → 12/5/03

Bibliographical note

Publisher Copyright:
© 2003 IEEE.

Keywords

Application software
Cities and towns
Computer science
Data engineering
Microprocessors
Optimizing compilers
Prefetching
Programming profession
Runtime
Software performance

Access

10.1109/MICRO.2003.1253194

OpenUrl availability

Full text

Cite this

Lu, J., Chen, H., Fu, R., Hsu, W. C., Othmer, B., Yew, P. C., & Chen, D. Y. (2003). The performance of runtime data cache prefetching in a dynamic optimization system. In Proceedings - 36th International Symposium on Microarchitecture, MICRO 2003 (pp. 180-190). Article 1253194 (Proceedings of the Annual International Symposium on Microarchitecture, MICRO; Vol. 2003-January). IEEE Computer Society. https://doi.org/10.1109/MICRO.2003.1253194

The performance of runtime data cache prefetching in a dynamic optimization system. / Lu, Jiwei; Chen, H.; Fu, Rao et al.
Proceedings - 36th International Symposium on Microarchitecture, MICRO 2003. IEEE Computer Society, 2003. p. 180-190 1253194 (Proceedings of the Annual International Symposium on Microarchitecture, MICRO; Vol. 2003-January).

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

Lu, J, Chen, H, Fu, R, Hsu, WC, Othmer, B, Yew, PC & Chen, DY 2003, The performance of runtime data cache prefetching in a dynamic optimization system. in Proceedings - 36th International Symposium on Microarchitecture, MICRO 2003., 1253194, Proceedings of the Annual International Symposium on Microarchitecture, MICRO, vol. 2003-January, IEEE Computer Society, pp. 180-190, 36th International Symposium on Microarchitecture, MICRO 2003, San Diego, United States, 12/3/03. https://doi.org/10.1109/MICRO.2003.1253194

Lu J, Chen H, Fu R, Hsu WC, Othmer B, Yew PC et al. The performance of runtime data cache prefetching in a dynamic optimization system. In Proceedings - 36th International Symposium on Microarchitecture, MICRO 2003. IEEE Computer Society. 2003. p. 180-190. 1253194. (Proceedings of the Annual International Symposium on Microarchitecture, MICRO). doi: 10.1109/MICRO.2003.1253194

@inproceedings{9159413e5ac047ff90cd2ca43ed47819,

title = "The performance of runtime data cache prefetching in a dynamic optimization system",

abstract = "Traditional software controlled data cache prefetching is often ineffective due to the lack of runtime cache miss and miss address information. To overcome this limitation, we implement runtime data cache prefetching in the dynamic optimization system ADORE (ADaptive Object code Reoptimization). Its performance has been compared with static software prefetching on the SPEC2000 benchmark suite. Runtime cache prefetching shows better performance. On an Itanium 2 based Linux workstation, it can increase performance by more than 20% over static prefetching on some benchmarks. For benchmarks that do not benefit from prefetching, the runtime optimization system adds only 1%-2% overhead. We have also collected cache miss profiles to guide static data cache prefetching in the ORC compiler. With that information the compiler can effectively avoid generating prefetches for loops that hit well in the data cache.",

keywords = "Application software, Cities and towns, Computer science, Data engineering, Microprocessors, Optimizing compilers, Prefetching, Programming profession, Runtime, Software performance",

author = "Jiwei Lu and H. Chen and Rao Fu and Hsu, {Wei Chung} and B. Othmer and Yew, {Pen Chung} and Chen, {Dong Yuan}",

note = "Publisher Copyright: {\textcopyright} 2003 IEEE.; 36th International Symposium on Microarchitecture, MICRO 2003 ; Conference date: 03-12-2003 Through 05-12-2003",

year = "2003",

doi = "10.1109/MICRO.2003.1253194",

language = "English (US)",

series = "Proceedings of the Annual International Symposium on Microarchitecture, MICRO",

publisher = "IEEE Computer Society",

pages = "180--190",

booktitle = "Proceedings - 36th International Symposium on Microarchitecture, MICRO 2003",

}

TY - GEN

T1 - The performance of runtime data cache prefetching in a dynamic optimization system

AU - Lu, Jiwei

AU - Chen, H.

AU - Fu, Rao

AU - Hsu, Wei Chung

AU - Othmer, B.

AU - Yew, Pen Chung

AU - Chen, Dong Yuan

PY - 2003

Y1 - 2003

N2 - Traditional software controlled data cache prefetching is often ineffective due to the lack of runtime cache miss and miss address information. To overcome this limitation, we implement runtime data cache prefetching in the dynamic optimization system ADORE (ADaptive Object code Reoptimization). Its performance has been compared with static software prefetching on the SPEC2000 benchmark suite. Runtime cache prefetching shows better performance. On an Itanium 2 based Linux workstation, it can increase performance by more than 20% over static prefetching on some benchmarks. For benchmarks that do not benefit from prefetching, the runtime optimization system adds only 1%-2% overhead. We have also collected cache miss profiles to guide static data cache prefetching in the ORC compiler. With that information the compiler can effectively avoid generating prefetches for loops that hit well in the data cache.

AB - Traditional software controlled data cache prefetching is often ineffective due to the lack of runtime cache miss and miss address information. To overcome this limitation, we implement runtime data cache prefetching in the dynamic optimization system ADORE (ADaptive Object code Reoptimization). Its performance has been compared with static software prefetching on the SPEC2000 benchmark suite. Runtime cache prefetching shows better performance. On an Itanium 2 based Linux workstation, it can increase performance by more than 20% over static prefetching on some benchmarks. For benchmarks that do not benefit from prefetching, the runtime optimization system adds only 1%-2% overhead. We have also collected cache miss profiles to guide static data cache prefetching in the ORC compiler. With that information the compiler can effectively avoid generating prefetches for loops that hit well in the data cache.

KW - Application software

KW - Cities and towns

KW - Computer science

KW - Data engineering

KW - Microprocessors

KW - Optimizing compilers

KW - Prefetching

KW - Programming profession

KW - Runtime

KW - Software performance

UR - http://www.scopus.com/inward/record.url?scp=67650020024&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67650020024&partnerID=8YFLogxK

U2 - 10.1109/MICRO.2003.1253194

DO - 10.1109/MICRO.2003.1253194

M3 - Conference contribution

AN - SCOPUS:67650020024

T3 - Proceedings of the Annual International Symposium on Microarchitecture, MICRO

SP - 180

EP - 190

BT - Proceedings - 36th International Symposium on Microarchitecture, MICRO 2003

PB - IEEE Computer Society

T2 - 36th International Symposium on Microarchitecture, MICRO 2003

Y2 - 3 December 2003 through 5 December 2003

ER -

The performance of runtime data cache prefetching in a dynamic optimization system

Abstract

Publication series

Other

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this