CSE 291C
Seminar on Heterogeneous Computing
Monday, Wednesday 10:00-11:20
AP&M 4218

Professor Fran Berman
AP&M 4202
berman@cs.ucsd.edu

The class will focus on current topics in Heterogeneous Computing or Metacomputing. Students will be responsible for presenting a paper of their choice relevant to metacomputing and completion of a project. Projects will be individualized to the student. All project proposals (a one page description of what you plan to do and project milestones throughout the quarter) are due on Wednesday, October 9.

Schedule

October 1: Berman lecture
October 2: Berman lecture
October 7: DARPA
October 9: Silvia Figueira
October 14: Neil Spring
October 16: Minnesota
October 21: Shelly Strout
October 23: SDSC
October 28: Gary Shao
October 30: Greg Bruno
November 4: Rich Wolski
November 6: Jenny Schopf
November 11: Veteran's Day
November 13: Marcio Faerman
November 18: Supercomputing 96
November 20: Supercomputing 96
November 25: Project presentations (Spring, Figueira)
December 2: Project presentations (Bruno, Shao)
December 3: Project presentations [2:00] (Schopf, Downey)
December 12: Project presentations [9:00] (Faerman, Strout)

Reading List

  • Rick Stevens' Keynote Address, 1996 Heterogeneous Computing Workshop
  • "Globus: A Metacomputing Infrastructure Toolkit", Ian Foster and Carl Kesselman, available at ftp://ftp.mcs.anl.gov/pub/nexus/reports/globus.ps.Z
  • "Legion -- A View from 50,000 ft." , Andrew Grimshaw
  • TR 96-3 A Federated Model for Scheduling in Wide-Area Systems by Jon B. Weissman and Andrew S. Grimshaw
  • TR 96-4 The Interference Paradigm for Network Job Scheduling by Jon B. Weissman
  • ECO: Efficient Collective Operations for Communication on Heterogeneous Networks Tech Report by Bruce Lowekamp and Adam Beguelin
  • DynamicPVM
  • MIST
  • DOME
  • Parform
  • AppLeS
  • MARS: A Framework for Minimizing the Job Execution Time in a Metacomputing Environment by Gehring and Reinfeld
  • Epema, D. H. J., Livny, M., van Dantzig, R., Evers, X. and Pruyne, J. A worldwide flock of Condors: Load sharing among workstation clusters. Future Generation Computer Systems. 12(1):53-65. May, 1996.
  • Condor.
  • Multimethod Communication for High-Performance Networked Computing Systems by I. Foster, J. Geisler, C. Kesselman and S. Tuecke
  • "I/O Requirements of Scientific Applications: An Evolutionary View" , Evgenia Smirni, Ruth A. Aydt, Andrew A. Chien, Daniel A. Reed - HPDC-96
  • "Input/Output Characteristics of Scalable Parallel Applications" , Phyllis E. Crandall Ruth A. Aydt Adrew A. Chien Daniel A. Reed - Supercomputing '95
  • "Scalable I/O Initiative (SIO)" , Caltech
  • "Dynamic I/O Characterization of I/O Intensive Scientific Applications" , Barbara K. Pasquale and George C. Polyzos, Proceedings of Supercomputing '94, Washington, D.C., November 14-18,
  • "A Static Analysis of I/O Characteristics of Scientific Applications in a Production Workload" , Barbara K. Pasquale and George C. Polyzos, Proceedings of Supercomputing '93, pp. 388-396, Portland, Oregon, November 1993.
  • Modeling the Performance of Parallel Programs, University of Rochester TR 589, by Wagner Meira Jr.
  • "Effective Distributed Scheduling of Parallel Workloads", Andrea C. Dusseau, Remzi H. Arpaci, David E. Culler - Sigmetrics'96 Conference on the Measurement and Modeling of Computers

    Links of Interest

  • UCSD Parallel Computation Laboratory
  • San Diego Supercomputer Center
  • AppLeS Home Page
  • Heterogeneous Computing at UCSD