PITTSBURGH-More than 80 people from academia, industry and government will be meeting to discuss "Dependability in Real Life" at the High Dependability Computing Consortium's (HDCC), second workshop, May 7-8, at Chaminade in Santa Cruz, Calif.
The HDCC was established in December 2000,by Carnegie Mellon University and NASA's Ames Research Center in the Silicon Valley.
The conference is co-chaired by Peter Norvig, chief, Computational Sciences Division, NASA Ames, Patrick Mantey, professor of computer engineering at the University of Southern California, and dependability expert Brian Reid, who recently became the first Carnegie Mellon faculty member to be headquartered in the Silicon Valley.
The HDCC's mission is to eliminate failures in computing systems critical to the welfare of society. To date, 14 information technology-related firms are working with Carnegie Mellon and NASA to establish a research agenda. They include Adobe Systems, Inc., Compaq Computer Corp., Five Nine Solutions, Inc., Hewlett-Packard Corp., IBM Corp., ILOG, Inc., Marimba, Inc., Microsoft Corp., Novell, Inc., Oracle Corp., SGI, Inc., Siebel Systems, Inc., Sybase, Inc., and Sun Microsystems, Inc.
The workshop will consist of a series of two-hour sessions that include a keynote talk about real-life dependability. Then, speaker and audience will interact in a question-and-answer period followed by breakout sessions from which each group will formulate research problems related to the speaker's talk.
The agenda includes:
Monday, May 7
- 8:30 a.m. James Gray, manager, Bay Area Research Center, Microsoft Research, "Internet Reliability."
- 11:15 a.m. Lynn Wheeler, chief technology officer, First Data Corp. "Engineering Real Systems to Be More Reliable."
- 2 p.m. Steven A. Gonzalez, branch chief for operations research and strategic development, NASA, Johnson Space Center.
- 4:30 p.m. Dependability Showcase and demonstrations by individuals and companies/cocktail hour.
Tuesday, May 8
- 8:30 a.m. Rohan Champion , chairman of the board and founder, eTime Capital and (formerly) Federal Express, "Reliability at FedEx and eTime Capital."
- 11:15 a.m. Wesley Sawyer, manager, high availability research and development, Hewlett-Packard Corp. "HP's 'Always On' Infrastructure."
- 2 p.m. Bruce Maggs, vice president research, Akamai Technologies and associate professor of computer science, Carnegie Mellon. "Challenges in Building a Reliable System of Tens of Thousands of Servers."
- 6 p.m. Dinner speaker, Harry MacDonald, director of NASA Ames.
"Dependable systems have become part of the landscape of modern life, but almost nothing about them exists in our research literature," says James H. Morris, dean of Carnegie Mellon's School of Computer Science. "Most observers agree that there is not much interaction between research in dependable computing and the actual technological systems for which dependability is important. This non-interaction goes in both directions: neither community is particularly aware of what the other is doing.
"We're hoping that we can meld NASA's tradition of 'failure is not an option' with the talent and energy of the computer industry into a consortium that benefits everyone. All of the communities involved, NASA, the aerospace companies, the Silicon Valley systems companies and the universities, have unique contributions to make."
For media wishing to attend the conference, contact Cleah Schlueter, School of Computer Science, Carnegie Mellon -Cleah@cs.cmu.edu.
For more information on the HDCC and the conference, check the Web site at http://www.hdcc.cs.cmu.edu.