MCAMP: communication optimization on massively parallel machines with hierarchical scratch-pad memory
Pages 102 - 111
Abstract
Massively parallel machines that integrate a large number of simple processors and small scratch-pad memories (SPMs) into a single chip can achieve a high peak performance per watt of power. In these machines, communication optimizations are important because the communication bandwidth tends to be a bottleneck. Previously proposed communication optimizations using copy candidates, which have been shown to be effective, detect frequently reused array regions by compile-time analysis and copy the regions to scratch-pad memories nearer to the processors. However, they have been proposed for uniprocessor systems or small parallel machines with one or more layers of scratch-pad memories, and the analysis time increases when they are applied to massively parallel machines. In this paper, we propose Multilayer Copy-candidate Analysis for Massively Parallel machines (MCAMP), a communication optimization method for massively parallel machines. MCAMP re-formalizes the framework used in earlier works and improves the scalability of the analysis by assuming the homogeneity of the target systems. We implemented an MCAMP optimizer, which takes an input program that consists of perfectly nested loops containing array references and computation codes, and generates optimized communication. We measured the performance of the output programs of the MCAMP optimizer by executing them on a real massively parallel machine GRAPE-DR using a software tool chain that we also implemented. We showed that MCAMP can achieve optimal data transfer patterns and comparable performance to that of hand-optimized codes with a short analysis time.
References
[1]
lp_solve. version 5.5.0.11 http://sourceforge.net/projects/lpsolve.
[2]
J. Absar and F. Catthoor. Reuse analysis of indirectly indexed arrays. ACM Trans. Des. Autom. Electron. Syst., 11(2):282--305, 2006.
[3]
P. Briggs, K. D. Cooper, and L. Torczon. Coloring register pairs. ACM Lett. Program. Lang. Syst., 1(1):3--13, 1992.
[4]
E. Brockmeyer, M. Miranda, H. Corporaal, and F. Catthoor. Layer assignment techniques for low energy in multi-layered memory organisations. In DATE '03: Proceedings of the conference on Design, Automation and Test in Europe, page 11070, Washington, DC, USA, 2003. IEEE Computer Society.
[5]
J. P. Diguet, S. Wuytack, F. Catthoor, and H. D. Man. Formalized methodology for data reuse exploration in hierarchical memory mappings. In ISLPED '97: Proceedings of the 1997 international symposium on Low power electronics and design, pages 30--35, New York, NY, USA, 1997. ACM.
[6]
S. Hiranandani, K. Kennedy, and C.-W. Tseng. Compiler optimizations for fortran D on MIMD distributed-memory machines. In Supercomputing '91: Proceedings of the 1991 ACM/IEEE conference on Supercomputing, pages 86--100, New York, NY, USA, 1991. ACM.
[7]
I. Issenin, E. Brockmeyer, B. Durinck, and N. Dutt. Multiprocessor system-on-chip data reuse analysis for exploring customized memory hierarchies. In DAC '06: Proceedings of the 43rd annual conference on Design automation, pages 49--52, New York, NY, USA, 2006. ACM.
[8]
I. Issenin, E. Brockmeyer, M. Miranda, and N. Dutt.Drdu: A data reuse analysis technique for efficient scratch-pad memory management. ACM Trans. Des. Autom. Electron. Syst., 12(2):15, 2007.
[9]
I. Issenin and N. Dutt. Data reuse driven energy-aware MPSoC co-synthesis of memory and communication architecture for streaming applications. In CODES+ISSS '06: Proceedings of the 4th international conference on Hardware/software codesign and system synthesis, pages 294--299, New York, NY, USA, 2006. ACM.
[10]
M. Kandemir and A. Choudhary. Compiler-directed scratch pad memory hierarchy design and management. In DAC '02: Proceedings of the 39th conference on Design automation, pages 628--633, New York, NY, USA, 2002. ACM.
[11]
J. Li and M. Chen. Compiling communication-efficient programs for massively parallel machines. IEEE Transactions on Parallel and Distributed Systems, 02(3):361--376, 1991.
[12]
J. Makino, K. Hiraki, and M. Inaba. GRAPE-DR: 2-Pflops massively-parallel computer with 512-core, 512-Gflops processor chips for scientific computing. In SC '07: Proceedings of the 2007 ACM/IEEE conference on Supercomputing, pages 1--11, New York, NY, USA, 2007. ACM.
[13]
O. Ozturk, M. Kandemir, M. J. Irwin, and S. Tosun. Multi-level on-chip memory hierarchy design for embedded chip multiprocessors. In ICPADS '06: Proceedings of the 12th International Conference on Parallel and Distributed Systems, pages 383--390, Washington, DC, USA, 2006. IEEE Computer Society.
[14]
T. Van Achteren, G. Deconinck, F. Catthoor, and R. Lauwereins. Data reuse exploration techniques for loop-dominated applications. In DATE '02: Proceedings of the conference on Design, automation and test in Europe, page 428, Washington, DC, USA, 2002. IEEE Computer Society.
Index Terms
- MCAMP: communication optimization on massively parallel machines with hierarchical scratch-pad memory
Recommendations
Experiences on Porting a Parallel Objects Environment from a Transputer Network to a PVM-Based System
PDP '96: Proceedings of the 4th Euromicro Workshop on Parallel and Distributed Processing (PDP '96)Abstract: Parallel Objects is a powerful model for distributed/parallel Object-Oriented programming. Goal of this paper is to present the approach adopted in porting the support of the Parallel Objects environment, originally implemented for a massively ...
DRDU: A data reuse analysis technique for efficient scratch-pad memory management
In multimedia and other streaming applications, a significant portion of energy is spent on data transfers. Exploiting data reuse opportunities in the application, we can reduce this energy by making copies of frequently used data in a small local ...
Dynamic scratch-pad memory management with data pipelining for embedded systems
Advanced Topics on Scalable ComputingIn this paper, we propose an effective data pipelining technique, SPDP (Scratch-Pad Data Pipelining), for dynamic scratch-pad memory (SPM) management with DMA (Direct Memory Access). Our basic idea is to overlap the execution of CPU instructions and DMA ...
Comments
Information & Contributors
Information
Published In
October 2008
328 pages
ISBN:9781605582825
DOI:10.1145/1454115
- General Chair:
- Andreas Moshovos,
- Program Chairs:
- David Tarditi,
- Kunle Olukotun
Copyright © 2008 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]
Sponsors
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 25 October 2008
Check for updates
Author Tags
Qualifiers
- Research-article
Conference
PACT '08: International Conference on Parallel Architectures and Compilation Techniques
October 25 - 29, 2008
Ontario, Toronto, Canada
Acceptance Rates
Overall Acceptance Rate 121 of 471 submissions, 26%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 247Total Downloads
- Downloads (Last 12 months)6
- Downloads (Last 6 weeks)1
Reflects downloads up to 03 Jan 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in