PRELIMINARY PROGRAM FOR STMCS'07

Session 1: PROGRAMMING MODELS, 1:35-2:45PM, Session Chair: Mary Hall
Toward a Toolchain for Pipeline Parallel Programming on CMPs
John Giacomoni, Tipp Moseley, Graham Price, Brian Bushnell, Manish Vachharajani, and Dirk Grunwald, University of Colorado
Paper: PDF
Presentation: PPT
Self-Describing Components: A Programming Model for Multicore Systems
James C. Browne and Nasim Mahmood, University of Texas
Paper: PDF
Presentation: PDF
Fuce machine and its programming system
S. AMAMIYA, Hiroshi FUJITA, Ryuzo HASEGAWA and Makoto AMAMIYA, Kyushu University
Paper: PDF
Presentation: PPT
Short Presentation: Automatic Application-Specific Customization of Multicore Processor Microarchitecture
Shobana Padmanabhan, Ron K. Cytron, and John W. Lockwood, Washington University in St. Louis
Paper: PDF
Presentation: PDF
Session 2: EXPERIENCE WITH EXISTING SYSTEMS, 2:45-3:25PM, Session Chair: Jim Dehnert
Developing and deploying software on the UltraSPARC-T1
Raj Prakash and Darryl Gove, Sun Microsystems Inc.
Paper: PDF
Presentation: PDF
An Integrated Tools Platform for Multi-Core Enablement
Beth Tibbitts, Evelyn Duesterwald, IBM T.J. Watson Research Center
Paper: PDF
Presentation: PPT
BREAK 3:30-4:00PM
Session 3: SOFTWARE SYSTEM ISSUES, 4:00-5:10PM, Session Chair: Manish Vachharajani
Position Paper - ToleRace: Tolerating and Detecting Races
Rahul Nagpaly, Karthik Pattabiramanz, Darko Kirovski, and Benjamin Zorn IISc Bangalore, University of Illinois at Urbana-Champaign, and Microsoft Research
Paper: PDF
Application-Specific System Customization on Many-Core Platforms: The VT-ASOS Framework
Godmar Back and Dimitrios S. Nikolopoulos, Virginia Tech
Paper: PDF
Presenation: PPT
Scalability : The Software Problem
Jonathan Appavoo, Volkmar Uhlig, and Dilma da Silva, IBM T. J. Watson Research Center
Paper: PDF
Presentation: PDF
Short Presentation: Heterogeneous Chip Multiprocessor Design for Virtual Machines
Dan Upton and Kim Hazelwood, University of Virginia
Paper: PDF
Presentation: PPT
DISCUSSION, Is it possible to make concurrent programming simple (or simple enough)? 5:10-5:30PM