Charles Koelbel

Personal Information

Business Address

Home Address

Career Goals

• Parallel and distributed computation, particularly hierarchical architectures
• High-level programming languages, particularly more abstract paradigms like object-oriented languages and domain-specific languages like MATLAB
• Compilers, interpreters, and tools to efficiently support those languages on the above machines
• Improved tools for interoperating between different languages, programming paradigms, and machines
• Improved user interfaces for high-level tasks, particularly data visualization and monitoring of computations

Education

• Purdue University, West Lafayette, IN - Ph.D. Computer Science - 1990
• Purdue University, West Lafayette, IN - M.S. Computer Science - 1985
• Augustana College, Rock Island, IL - B.A. Mathematics and Computer Science - 1983

Work Experience

Rice University, Center for Research on Parallel Computation

August 1990-present

Most recent job title: Research Scientist II (not tenure track)

Supervisor: Ken Kennedy

CRPC is an NSF-funded Science and Technology Center whose mission is to make parallel computation truly usable for scientists and engineers. I worked as a researcher in several CRPC projects, including publishing technical papers (including one book), supervising students, researchers and other workers, managing grants in excess of $300,000, and submitting proposals. The individual projects are listed below.

Research Experience

National Computational Science Alliance (Rice University, 1997-present)

Program to to maintain world leadership in computational science and engineering by:

• providing access to high-end computing infrastructure for the academic scientific and engineering community;
• partnering with universities, states, and industry to facilitate and enhance that access;
• supporting the effective use of such infrastructure through training, consulting, and related support services;
• being a vigorous early user of experimental and emerging high performance technologies that offer high potential for advancing computational science and engineering;
• facilitating the development of the intellectual capital required to maintain world leadership.

Rice University is involved in all parts of the Alliance: Enabling Technologies (ET); Application Technologies (AT); Education, Outreach, and Training (EOT); and Partners for Advanced Computational Services (PACS). I am directly involved in the ET "Parallel Computing" team led by Ken Kennedy and in the PACS "Technology Deployment Partners" thrust. In the parallel computing team, I develop compilers for HPF, manage research funds at Rice University, coordinate activites between the other team members, and often represent the team to other parts of the Alliance (particularly the Executive Committee). In the technology deployment role, my duties are similar; the major difference is that the technical activities are directed at transferring the technology to scientific users rather than development of a specific piece of software.

DOD High Performance Computing Modernization Program (Rice University, 1996-present)

Project to rapidly upgrade the computing capacity of the Department of Defense, both in hardware and software capabilities. Rice takes part in the Programming Environment and Training program, a comprehensive technology transfer program that aims to bring leading-edge university research to bear on DOD needs. We manage the programming tools activities at three of the DOD Major Shared Resource Centers.

D System Programming Tools (Rice University, 1992-Present)

Project to design and implement a set of tools for programming in data-parallel languages such as HPF and Fortran D. The key issues include providing feedback in terms of the original source rather than the compiler-transformed behavior.

Scalable Input/Output Consortium (Rice University, 1994-1997)

Project to systematically investigate a primary obstacle to effective use of current and future high-performance computing systems: getting data into, around, and out of the system. Technical issues (for the Rice effort) were automatically translating high-level constructs for I/O into efficient low-level operations.

Fortran D Programming Language (Rice University, 1990-1995)

Project to design and implement a set of FORTRAN extensions for machine-independent parallel programming. Subsidiary goals included high execution efficiency (compared with hand-coded programs) and programming environment support.

Kali Programming Language (Purdue University, 1988-1990}

Project to design and implement a language for scientific computing on non-shared memory multi-processors for use as a testbed for new compilation techniques on those machines. Target architectures included the NCUBE/7 and Intel iPSC/2 hypercubes.

BLAZE Programming Language (Purdue University, 1985-1988)

Predecessor of Kali project which designed a compiler to detect parallelism in sequential code and exploit this with optimizing transformations. Target architectures included the IBM RP3 and Sequent multi-processors.

Professional Activities

• High Performance Fortran Forum
  • Executive Director, 1992-1993
  • Active member, 1993-1997
• High Performance Debugger Forum - member, 1997
• Reviewer
  • IEEE Computer
  • IEEE Parallel and Distributed Computing
  • IEEE Software
  • Concurrency: Practice and Experience
  • Journal of Parallel Programming
• Program Committee Member
  • 1999 SC'99 Conference HPC Challenge Chair (Portland, OR)
  • 1999 SIAM Conference on Parallel Processing (San Antonio, TX)
  • 1999 IEEE Symposium on the Frontiers of Massively Parallel Processing (Annapolis, MD)
  • 1998 SC'98 Conference (Orlando, FL)
  • 1998 HPF Users Group Meeting (Porto, Portugal)
  • 1998 DOD High Performance Computing Modernization Program (Houston, TX)
  • 1997 SIAM Conference on Parallel Processing (Minneapolis, MN)
  • 1997 HPF Users Group Meeting (Santa Fe, NM)
  • 1996 Supercomputing '96 (Pittsburgh, PA)
  • 1995 Principles and Practice of Paralel Programming (Santa Barbara)
  • Workshop on Languages, Compilers, and Run-Time Systems for Scalable Computers (May 1995, Troy, NY)
  • 1995 Frontiers of Massively Parallel Computation (McLean, VA)
  • 1994 Scalable High Performance Computing Conference (Knoxville, TN)
  • 1993 International Conference on Supercomputing (Tokyo)
  • 1992 International Conference on Supercomputing (Washington, DC)
  • Workshop on Languges, Compilers and Run-time Environments for Distributed Memory Machines (Boulder, CO Oct. 1992)

Publications