Compiler optimizations for Java aglets in distributed data intensive applications

ABSTRACT

Code migration in light of distributed data intensive computing poses interesting compilation issues. In this work, we first define a small extension to the aglet model to allow data distribution. In our aglet program, data is distributed over the network using annotations (this is similar to High Performance Fortran (HPF) where the programmer specifies data distributions through annotations). We analyze the program using the annotations and use the 'owner computes' rule to determine where a given computation should take place. The compiler then schedules the aglet through the network and also determines the data it should carry during its migration. Determining efficient schedule of the aglet and which data to carry during migration poses interesting issues.We propose two strategies to optimize the aglet schedule. The first strategy called Take All Live Data: (TALD) attempts to carry all the live definitions of variables from a given node when visited. The second strategy Take Only Needed Data (TOND) attempts to carry only those definitions whose uses are in the destination node. The goal of the first strategy is to minimize the number of migrations which are expensive due to high serialization overheads The second strategy aims to minimize bandwidth consumption during a migration. This could significantly reduce the communication overhead due to minimal amount of data carried during each migration. We have implemented both the strategies in the Jikes compiler from IBM. We have evaluated it on a distributed database application and show benefits of both the strategies on large and small databases. The results show that strategies generated by our compiler analysis reduce the overheads and improve execution time.

REFERENCES

• 1: Antonio Carzaniga , Gian Pietro Picco , Giovanni Vigna, Designing distributed applications with mobile code paradigms, Proceedings of the 19th international conference on Software engineering, p.22-32, May 17-23, 1997, Boston, Massachusetts, United States [doi>10.1145/253228.253236]
• 2: Iqbal A, Baumann J, and Straβer M. Efficient Algorithms to Find Optimal Agent Migration Strategies. Technical report, IPVR, University of Stuttgart, 1998.
• 3: Mobile Agent Programming in Ajanta, Proceedings of the 19th IEEE International Conference on Distributed Computing Systems, p.190, May 31-June 04, 1999
• 4: Venners B. Under The Hood: The Architecture of Aglets. http://www.javaworld.com/javaworld/jw-04-1997/jw-04-hood.html, 1997.
• 5: Brian Wims , Cheng-Zhong Xu, TRAVELER: A Mobile Agent Based Infrastructure for Wide Area Parallel Computing, Proceedings of the First International Symposium on Agent Systems and Applications Third International Symposium on Mobile Agents, p.258, October 03-06, 1999
• 6: William R. Cockayne , Michael Zyda, Mobile Agents, Manning Pu, 1997
• 7: Chess D, Harrison C, and Kershenbaum A. Mobile Agents: Are they a good idea? In Research Report. IBM Research Division, 1994.
• 8: Danny B. Lange , Oshima Mitsuru, Programming and Deploying Java Mobile Agents Aglets, Addison-Wesley Longman Publishing Co., Inc., Boston, MA, 1998
• 9: Daniel Wu , Divyakant Agrawal , Amr El Abbadi, StratOSphere: mobile processing of distributed objects in Java, Proceedings of the 4th annual ACM/IEEE international conference on Mobile computing and networking, p.121-132, October 25-30, 1998, Dallas, Texas, United States [doi>10.1145/288235.288267]
• 10: Jonathan Bredin , David Kotz , Daniela Rus, Utility Driven Mobile-Agent Scheduling, Dartmouth College, Hanover, NH, 2005
• 11: Jaspal Subhlok , Peter Lieu , Bruce Lowekamp, Automatic node selection for high performance applications on networks, Proceedings of the seventh ACM SIGPLAN symposium on Principles and practice of parallel programming, p.163-172, May 04-06, 1999, Atlanta, Georgia, United States
• 12: Luca Cardelli, Abstractions for mobile computations, Secure Internet programming: security issues for mobile and distributed objects, Springer-Verlag, London, 2001
• 13: Daniel Hagimont , L. Ismail, A performance evaluation of the mobile agent paradigm, Proceedings of the 14th ACM SIGPLAN conference on Object-oriented programming, systems, languages, and applications, p.306-313, November 01-05, 1999, Denver, Colorado, United States
• 14: Straβer M., Baumann J., and Hohl F. Mole - A Java Based Mobile Agent System. In Special Issues in Object-Oriented Programming, Workshop Reader ECOOP'96, dpunkt.verlag, pages 327-334, 1996.
• 15: Straβer M and Schwehm M. A Performance Model for Mobile Agent Systems. In Distributed Processing Techniques and Applications DPPTA '97, pages 1132-1140, 1997.
• 16: M. Tamer Özsu , Patrick Valduriez, Principles of distributed database systems (2nd ed.), Prentice-Hall, Inc., Upper Saddle River, NJ, 1999

INDEX TERMS

Primary Classification:
  D. Software
  D.3 PROGRAMMING LANGUAGES
      D.3.4 Processors
          Subjects: Optimization

Additional Classification:
  D. Software
  D.1 PROGRAMMING TECHNIQUES
      D.1.3 Concurrent Programming
          Subjects: Distributed programming
  D.3 PROGRAMMING LANGUAGES
      D.3.2 Language Classifications

          Nouns: Java
      D.3.4 Processors
          Subjects: Compilers

General Terms:
Algorithms, Experimentation, Performance, Theory