High-performance computing pioneer Paul Messina has been named an Argonne National Laboratory Distinguished Fellow, the laboratory’s highest scientific and engineering rank. Comparable in stature to an endowed chair at a top-ranked university, the Argonne Distinguished Fellow title rank recognizes sustained outstanding scientific and engineering research and can also be associated with outstanding technical leadership of major, complex, high-priority projects.
Paul, an applied mathematician and computer scientist, first joined Argonne’s Applied Mathematics Division in 1973 to manage the Argonne Subroutine Library and to found and lead the User Services Group. In 1982, he served as founding director of Argonne’s Mathematics and Computer Science Division, a position he held until 1987. In 1983, Paul and his colleagues Jack Dongarra, Danny Sorensen, and Rusty Lusk established the parallel computing program at Argonne – nearly 10 years before computational science was broadly recognized and federally funded as a new paradigm for scientific investigation. Paul worked at Caltech from 1987-2002 and rejoined Argonne in 2002 as a Senior Computer Scientist. Since 2008, Paul has served as the Argonne Leadership Computing Facility’s (ALCF’s) Director of Science, where he helps attract top scientists to the ALCF user community.
“Paul holds a key scientific role that has the responsibility of charting the scientific directions of the ALCF, engaging the user community to refine the facility roadmap, and ensuring that the ALCF remains a world-leading capability for science,” said Associate Laboratory Director for Computing, Environment and Life Sciences Rick Stevens.
Over the course of Paul’s career, he has designed, directed, and otherwise executed numerous initiatives that have influenced U.S. policy and programs. He has also organized numerous workshops whose resulting reports have provided critical input to decision makers for over three decades. Paul’s extensive list of contributions to the field include:
- In the late 1980s throughout the 1990s, Paul wrote expository articles and book chapters on advanced computer architectures considered to be accessible explanations of high-performance scientific computer architectures and used as references in university courses.
- In 1990, Paul conceived and led the nation’s first parallel computing consortium, which led to the creation of the first highly parallel computer more powerful than vector machines for a wide variety of science and engineering applications. This effort was instrumental in planning for what became the federal High Performance Computing and Communications (HPCC) Program.
- Paul’s role in advancing computing infrastructure – an essential companion effort to high-performance computing hardware and software development efforts – includes his leadership of the CASA Gigabit Network Testbed, which created the world’s fastest network and proved the power and promise of distributing applications across wide area networks. As a component of the DARPA and National Science Foundation’s Gigabit Testbed Project, Paul’s work helped to demonstrate the technical feasibility of gigabit networking and explored the associated organizational and operational issues. Overall, this project laid critical parts of the foundation for today’s Internet. It also inspired several large grid research efforts in the U.S., such as NSF’s Teragrid Project, of which he was a co-PI, and Globus, and others internationally, that persist in support of large, community-driven, distributed scientific research.
- Paul led the Department of Energy’s Accelerated Strategic Computing Initiative from 1999 to 2000, which at the time was the world’s largest and most comprehensive scientific computing project.
- Paul was co-PI of the National Virtual Observatory Project from 2001 to 2002—a seminal project focused on the organization and mining of multiple scientific databases, illustrating the importance of obtaining new scientific knowledge in large, stored data for the first time on a very large scale.
- As Senior Advisor to Director General of CERN in 2002, Paul formulated the strategy and technical framework for acquiring, storing, and analyzing Large Hadron Collider data.
At the ALCF, Paul has been instrumental in the government’s planning efforts to develop an exascale system, organizing workshops on behalf of the DOE’s Office of Advanced Scientific Computing Research to identify the potential scientific and national security benefits of these next-generation systems, as well as the technical challenges that must be met. Since 2013, Paul has run the Argonne Training Program on Extreme-Scale Computing, a two-week intensive training course aimed at preparing the next generation of supercomputer users.
“Paul’s revolutionary work provided the foundation for so many scientific advances made at Argonne and across the world,” said ALCF Director Michael Papka. “He continues to be an important voice in growing the HPC community and mentoring future generations of computer and computational scientists.”