Projects Members Contact Us Back to HiPerSoft Home Page
Projects Visit Rice University

CGrADS Site Visit Participant - Daniel Chavarria


Educational Background

  • Undergraduate: B.S. in Computer Science (with honors), Universidad de Costa Rica, San José, Costa Rica, March 1994.
  • Graduate: M.S. in Computer Science (with honors), Instituto Tecnológico y de Estudios Superiores de Monterrey (ITESM), Monterrey, Nuevo León, Mexico, June 1998.
  • Fourth year of Rice University's Computer Science Ph.D. program, August 1998 - present. Advisor: Dr. John Mellor-Crummey.
Daniel Chavarria
Graduate Student, Computer Science
Rice University

Phone: (713)-348-3827
danich@cs.rice.edu
http://www.cs.rice.edu/~danich
Research Advisor: John Mellor-Crummey

Research Experience:

  • Currently working on optimizing generated code for HPF applications.
  • Summer 2000: Intern at Los Alamos National Laboratory's Advanced Computing Laboratory (ACL) . Worked on extending the SMARTS runtime system for parallel computing, to use it for non-numeric algorithms and mixed-mode applications (shared memory and message passing) (work done with Dr. Suvas Vajracharya). I also implemented a C++ class for launching applications securely on a Linux compute cluster, using the SSH protocol (work done with Ronald Minnich).
  • 1999-2000: Worked with Dr. John Mellor-Crummey and Dr. Vikram Adve on the implementation of multipartitioning within Rice's dHPF compiler.
  • 1997-1998: Worked with Dr. David Garza on the implementation of a novel data partitioning technique, called "Snake Partitioning" within a loop restructuring tool we implemented at ITESM.

Publications:

  1. Chavarría-Miranda, Daniel, John Mellor-Crummey and Trushar Sarang. "Data-Parallel Compiler Support for Multipartitioning", in proceedings of Euro-Par 2001 European Conference on Parallel Computing, Manchester, UK, August 2001.
  2. Darte, Alain, Daniel Chavarría-Miranda, Robert Fowler and John Mellor-Crummey. "On Efficient Parallelization of Line-Sweep Computations", in informal proceedings of CPC2001: Ninth International Workshop on Compilers for Parallel Computers, Edinburgh, Scotland UK, June 2001.
  3. Vajracharya, Suvas and Daniel Chavarría-Miranda. "Asynchronous Resource Management", in Proceedings of IPDPS01: International Parallel and Distributed Processing Symposium, San Francisco, California, April 2001.
  4. Broom, Bradley, Daniel Chavarría-Miranda, Guohua Jin, Rob Fowler, Ken Kennedy, John Mellor-Crummey and Qing Yi. "Overpartitioning with the Rice dHPF Compiler", in informal proceedings of HUG2000: The 4th Annual HPF User Group Meeting. Tokyo, Japan, October 2000.
  5. Chavarría-Miranda, Daniel and John Mellor-Crummey. "Towards Compiler Support for Scalable Parallelism using Multipartitioning", in proceedings of LCR2000: Fifth Workshop on Languages, Compilers, and Run-time Systems for Scalable Computers, Rochester, New York, May 2000.
  6. Chavarría-Miranda, Daniel. "Ejecución Eficiente de Ciclos Particionados en Forma de "Serpientes" en una Computadora de Memoria Distribuida", ("Efficient Execution of "Snake" Partitioned Loops on a Distributed Memory Multicomputer") Master's Thesis (in Spanish). Instituto Tecnológico y de Estudios Superiores de Monterrey (ITESM), Monterrey, Nuevo León, Mexico, May 1998.
  7. Garza-Salazar, David and Daniel Chavarría-Miranda. "On the Performance of Snake Partitioning: A Data Decomposition Technique that Reduces Communication and Exploits Locality" , in workshop on High Performance Computing in Developing Countries, EuroPar'97 Parallel Processing Conference. Passau, Germany, August 1997.

    Top


Awards and Honors:

  • June 1998: Honors Graduation, with Mention of Excellence, Master of Science in Computer Science, Instituto Tecnológico y de Estudios Superiores de Monterrey (ITESM). Highest GPA of the CS graduating class.
  • March 1994: Honors Graduation, Bachelor of Science in Computer Science, Universidad de Costa Rica. Highest GPA of the CS graduating class.
  • Student Member of the Association for Computing Machinery (ACM) since 1997.

Top