Strategies for dynamic memory allocation in hybrid architectures

ABSTRACT

Hybrid architectures combining the strengths of general-purpose processors with application-specific hardware accelerators can lead to a significant performance improvement. Our hybrid architecture uses a Java Virtual Machine as an abstraction layer to hide the complexity of the hardware/software interface between processor and accelerator from the programmer. The data communication between the accelerator and the processor often incurs a significant cost, which sometimes annihilates the original speedup obtained by the accelerator. This article shows how we minimise this communication cost by dynamically chosing an optimal data layout in the Java heap memory which is distributed over both the accelerator and the processor memory. The proposed self-learning memory allocation strategy finds the optimal location for each Java object's data by means of runtime profiling. The communication cost is effectively reduced by up to 86% for the benchmarks in the DaCapo suite (51% on average).

REFERENCES

• 1: Stephen M. Blackburn , Robin Garner , Chris Hoffmann , Asjad M. Khang , Kathryn S. McKinley , Rotem Bentzur , Amer Diwan , Daniel Feinberg , Daniel Frampton , Samuel Z. Guyer , Martin Hirzel , Antony Hosking , Maria Jump , Han Lee , J. Eliot B. Moss , B. Moss , Aashish Phansalkar , Darko Stefanović , Thomas VanDrunen , Daniel von Dincklage , Ben Wiedermann, The DaCapo benchmarks: java benchmarking development and analysis, Proceedings of the 21st annual ACM SIGPLAN conference on Object-oriented programming systems, languages, and applications, October 22-26, 2006, Portland, Oregon, USA
• 2: Andrew Borg , Rui Gao , Neil Audsley, A co-design strategy for embedded Java applications based on a hardware interface with invocation semantics, Proceedings of the 4th international workshop on Java technologies for real-time and embedded systems, October 11-13, 2006, Paris, France [doi>10.1145/1167999.1168010]
• 3: Edwin Hakkennes , Stamatis Vassiliadis, Multimedia Execution Hardware Accelerator, Journal of VLSI Signal Processing Systems, v.28 n.3, p.221-234, July 2001 [doi>10.1023/A:1011117608815]
• 4: Rachid Helaihel , Kunle Olukotun, Java as a specification language for hardware-software systems, Proceedings of the 1997 IEEE/ACM international conference on Computer-aided design, p.690-697, November 09-13, 1997, San Jose, California, United States
• 5: E. Lattanzi et al. Improving Java performance using dynamic method migration on FPGAs. International Journal of Embedded Systems, 1(3):228--236, 2005.
• 6: Roman Lysecky , Greg Stitt , Frank Vahid, Warp Processors, ACM Transactions on Design Automation of Electronic Systems (TODAES), v.11 n.3, p.659-681, July 2006
• 7: Rahul P. Maddimsetty , Jeremy Buhler , Roger D. Chamberlain , Mark A. Franklin , Brandon Harris, Accelerator design for protein sequence HMM search, Proceedings of the 20th annual international conference on Supercomputing, June 28-July 01, 2006, Cairns, Queensland, Australia [doi>10.1145/1183401.1183442]
• 8: Elena Moscu Panainte , Koen Bertels , Stamatis Vassiliadis, The Molen compiler for reconfigurable processors, ACM Transactions on Embedded Computing Systems (TECS), v.6 n.1, February 2007 [doi>10.1145/1210268.1210274]
• 9: C. Porthouse. Jazelle for execution environments. ARM Whitepaper, available online, May 2005.
• 10: Martin Schoeberl, A Java processor architecture for embedded real-time systems, Journal of Systems Architecture: the EUROMICRO Journal, v.54 n.1-2, p.265-286, January, 2008 [doi>10.1016/j.sysarc.2007.06.001]
• 11: Stamatis Vassiliadis , Stephan Wong , Georgi Gaydadjiev , Koen Bertels , Georgi Kuzmanov , Elena Moscu Panainte, The MOLEN Polymorphic Processor, IEEE Transactions on Computers, v.53 n.11, p.1363-1375, November 2004 [doi>10.1109/TC.2004.104]

INDEX TERMS

Primary Classification:
  B. Hardware
  B.3 MEMORY STRUCTURES
      B.3.2 Design Styles
          Subjects: Shared memory

Additional Classification:
  D. Software
  D.3 PROGRAMMING LANGUAGES
      D.3.4 Processors
          Subjects: Memory management (garbage collection)

General Terms:
Algorithms, Experimentation, Performance

Keywords:
hardware acceleration, java, memory management