Morphable cache architectures: Potential benefits

I. Kadayif, Mahmut Kandemir, Vijaykrishnan Narayanan, M. J. Irwin, J. Ramanujam

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

Abstract

Computer architects ha ve tried to mitigate the consequences of high memory latencies using a variety techniques. An example of these techniques is m ulti-lev elcaches to coun-teract the latency that results from having a memory that is slow er than the processor. Recent research has demon-strated that compiler optimizations that modify data lay-outs and restructure computation can be successful in im-proving memory system performance. However, in many cases, w orking with a fixed cache configuration prevents the application/compiler from obtaining the maximum perfor-mance. In addition, prompted by demands in portabilit y, long battery life, and low-cost packaging, the computer in-dustry has started viewing energy and pow er as decisive de-sign factors, along with performance and cost. This makes the job of the compiler/user even more dificult as one needs to strik e a balance betw een low pow er/energy consumption and high performance. Consequently, adapting the code to the underlying cache/memory hierarchy is becoming more and more dificult. In this paper, we tak e an alternate approach and attempt to adapt the cache arc hitecture to the softw are needs. We focus on array-dominated applications and measure the po-tential benefits that could be gained from a morphable (re-configurable) cache architecture. Our results show that not only difierent applications w ork best with difierent cache configurations, but also that difierent loop nests in a given application demand difierent configurations. Our results also indicate that the most suitable cache configuration for a giv en application or a single nest depends strongly on the ob-jectiv e function being optimized. For example, minimizing cache memory energy requires a difierent cache configuration for each nest than an objective which tries to minimize the overall memory system energy. Based on our experiments, w e conclude that fine-grain (loop nest-level) cache configu-ration management is an important step for a solution to the challenging architecture/softw are tradeofus aw aiting system designers in the future.

Original languageEnglish (US)
Title of host publicationProceedings of the 2001 ACM SIGPLAN Workshop on Optimization of Middleware and Distributed Systems, OM 2001
PublisherAssociation for Computing Machinery, Inc
Pages128-137
Number of pages10
ISBN (Print)1581134266, 9781581134261
DOIs
StatePublished - Aug 1 2001
Event2001 ACM SIGPLAN Workshop on Optimization of Middleware and Distributed Systems, OM 2001 - Snowbird, United States
Duration: Jun 18 2001 → …

Other

Other2001 ACM SIGPLAN Workshop on Optimization of Middleware and Distributed Systems, OM 2001
CountryUnited States
CitySnowbird
Period6/18/01 → …

Fingerprint

Data storage equipment
Cache memory
Costs
Packaging
Energy utilization
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Software

Cite this

Kadayif, I., Kandemir, M., Narayanan, V., Irwin, M. J., & Ramanujam, J. (2001). Morphable cache architectures: Potential benefits. In Proceedings of the 2001 ACM SIGPLAN Workshop on Optimization of Middleware and Distributed Systems, OM 2001 (pp. 128-137). Association for Computing Machinery, Inc. https://doi.org/10.1145/384198.384215
Kadayif, I. ; Kandemir, Mahmut ; Narayanan, Vijaykrishnan ; Irwin, M. J. ; Ramanujam, J. / Morphable cache architectures : Potential benefits. Proceedings of the 2001 ACM SIGPLAN Workshop on Optimization of Middleware and Distributed Systems, OM 2001. Association for Computing Machinery, Inc, 2001. pp. 128-137
@inproceedings{090c97036e2a49fab3a965f19bd2b132,
title = "Morphable cache architectures: Potential benefits",
abstract = "Computer architects ha ve tried to mitigate the consequences of high memory latencies using a variety techniques. An example of these techniques is m ulti-lev elcaches to coun-teract the latency that results from having a memory that is slow er than the processor. Recent research has demon-strated that compiler optimizations that modify data lay-outs and restructure computation can be successful in im-proving memory system performance. However, in many cases, w orking with a fixed cache configuration prevents the application/compiler from obtaining the maximum perfor-mance. In addition, prompted by demands in portabilit y, long battery life, and low-cost packaging, the computer in-dustry has started viewing energy and pow er as decisive de-sign factors, along with performance and cost. This makes the job of the compiler/user even more dificult as one needs to strik e a balance betw een low pow er/energy consumption and high performance. Consequently, adapting the code to the underlying cache/memory hierarchy is becoming more and more dificult. In this paper, we tak e an alternate approach and attempt to adapt the cache arc hitecture to the softw are needs. We focus on array-dominated applications and measure the po-tential benefits that could be gained from a morphable (re-configurable) cache architecture. Our results show that not only difierent applications w ork best with difierent cache configurations, but also that difierent loop nests in a given application demand difierent configurations. Our results also indicate that the most suitable cache configuration for a giv en application or a single nest depends strongly on the ob-jectiv e function being optimized. For example, minimizing cache memory energy requires a difierent cache configuration for each nest than an objective which tries to minimize the overall memory system energy. Based on our experiments, w e conclude that fine-grain (loop nest-level) cache configu-ration management is an important step for a solution to the challenging architecture/softw are tradeofus aw aiting system designers in the future.",
author = "I. Kadayif and Mahmut Kandemir and Vijaykrishnan Narayanan and Irwin, {M. J.} and J. Ramanujam",
year = "2001",
month = "8",
day = "1",
doi = "10.1145/384198.384215",
language = "English (US)",
isbn = "1581134266",
pages = "128--137",
booktitle = "Proceedings of the 2001 ACM SIGPLAN Workshop on Optimization of Middleware and Distributed Systems, OM 2001",
publisher = "Association for Computing Machinery, Inc",

}

Kadayif, I, Kandemir, M, Narayanan, V, Irwin, MJ & Ramanujam, J 2001, Morphable cache architectures: Potential benefits. in Proceedings of the 2001 ACM SIGPLAN Workshop on Optimization of Middleware and Distributed Systems, OM 2001. Association for Computing Machinery, Inc, pp. 128-137, 2001 ACM SIGPLAN Workshop on Optimization of Middleware and Distributed Systems, OM 2001, Snowbird, United States, 6/18/01. https://doi.org/10.1145/384198.384215

Morphable cache architectures : Potential benefits. / Kadayif, I.; Kandemir, Mahmut; Narayanan, Vijaykrishnan; Irwin, M. J.; Ramanujam, J.

Proceedings of the 2001 ACM SIGPLAN Workshop on Optimization of Middleware and Distributed Systems, OM 2001. Association for Computing Machinery, Inc, 2001. p. 128-137.

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

TY - GEN

T1 - Morphable cache architectures

T2 - Potential benefits

AU - Kadayif, I.

AU - Kandemir, Mahmut

AU - Narayanan, Vijaykrishnan

AU - Irwin, M. J.

AU - Ramanujam, J.

PY - 2001/8/1

Y1 - 2001/8/1

N2 - Computer architects ha ve tried to mitigate the consequences of high memory latencies using a variety techniques. An example of these techniques is m ulti-lev elcaches to coun-teract the latency that results from having a memory that is slow er than the processor. Recent research has demon-strated that compiler optimizations that modify data lay-outs and restructure computation can be successful in im-proving memory system performance. However, in many cases, w orking with a fixed cache configuration prevents the application/compiler from obtaining the maximum perfor-mance. In addition, prompted by demands in portabilit y, long battery life, and low-cost packaging, the computer in-dustry has started viewing energy and pow er as decisive de-sign factors, along with performance and cost. This makes the job of the compiler/user even more dificult as one needs to strik e a balance betw een low pow er/energy consumption and high performance. Consequently, adapting the code to the underlying cache/memory hierarchy is becoming more and more dificult. In this paper, we tak e an alternate approach and attempt to adapt the cache arc hitecture to the softw are needs. We focus on array-dominated applications and measure the po-tential benefits that could be gained from a morphable (re-configurable) cache architecture. Our results show that not only difierent applications w ork best with difierent cache configurations, but also that difierent loop nests in a given application demand difierent configurations. Our results also indicate that the most suitable cache configuration for a giv en application or a single nest depends strongly on the ob-jectiv e function being optimized. For example, minimizing cache memory energy requires a difierent cache configuration for each nest than an objective which tries to minimize the overall memory system energy. Based on our experiments, w e conclude that fine-grain (loop nest-level) cache configu-ration management is an important step for a solution to the challenging architecture/softw are tradeofus aw aiting system designers in the future.

AB - Computer architects ha ve tried to mitigate the consequences of high memory latencies using a variety techniques. An example of these techniques is m ulti-lev elcaches to coun-teract the latency that results from having a memory that is slow er than the processor. Recent research has demon-strated that compiler optimizations that modify data lay-outs and restructure computation can be successful in im-proving memory system performance. However, in many cases, w orking with a fixed cache configuration prevents the application/compiler from obtaining the maximum perfor-mance. In addition, prompted by demands in portabilit y, long battery life, and low-cost packaging, the computer in-dustry has started viewing energy and pow er as decisive de-sign factors, along with performance and cost. This makes the job of the compiler/user even more dificult as one needs to strik e a balance betw een low pow er/energy consumption and high performance. Consequently, adapting the code to the underlying cache/memory hierarchy is becoming more and more dificult. In this paper, we tak e an alternate approach and attempt to adapt the cache arc hitecture to the softw are needs. We focus on array-dominated applications and measure the po-tential benefits that could be gained from a morphable (re-configurable) cache architecture. Our results show that not only difierent applications w ork best with difierent cache configurations, but also that difierent loop nests in a given application demand difierent configurations. Our results also indicate that the most suitable cache configuration for a giv en application or a single nest depends strongly on the ob-jectiv e function being optimized. For example, minimizing cache memory energy requires a difierent cache configuration for each nest than an objective which tries to minimize the overall memory system energy. Based on our experiments, w e conclude that fine-grain (loop nest-level) cache configu-ration management is an important step for a solution to the challenging architecture/softw are tradeofus aw aiting system designers in the future.

UR - http://www.scopus.com/inward/record.url?scp=85053418229&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053418229&partnerID=8YFLogxK

U2 - 10.1145/384198.384215

DO - 10.1145/384198.384215

M3 - Conference contribution

AN - SCOPUS:85053418229

SN - 1581134266

SN - 9781581134261

SP - 128

EP - 137

BT - Proceedings of the 2001 ACM SIGPLAN Workshop on Optimization of Middleware and Distributed Systems, OM 2001

PB - Association for Computing Machinery, Inc

ER -

Kadayif I, Kandemir M, Narayanan V, Irwin MJ, Ramanujam J. Morphable cache architectures: Potential benefits. In Proceedings of the 2001 ACM SIGPLAN Workshop on Optimization of Middleware and Distributed Systems, OM 2001. Association for Computing Machinery, Inc. 2001. p. 128-137 https://doi.org/10.1145/384198.384215