DOE High Performance Computing Resource Providers Budget Code: KJ0102, KJ3501
Support the operation of High Performance Computing Resource Providers which provide critical resources for enabling Grand Challenge applications. To this end, the HPCRP's acquire large, full-scale, high performance computers exploiting a variety architectures for use by computational scientists working on Grand Challenge applications. The centers integrate these early high-performance computing systems into prototype heterogeneous computing configurations and make them available through the NREN. This allows for a wide spectrum of experiments for scalability studies, as well as an opportunity to provide Grand Challenge researchers access to the largest possible advanced systems.

The range of architectures includes Intel Paragon, IBM SP2, Cray T3E, and advanced networks of SMP's
Budget ($ M)
FY 95 Act 12.20
FY 96 Pres 12.39
FY 96 Est 12.39
FY 97 Rqst 15.63
Program Component Areas
  FY 96 FY 97
HECC 12.39 15.63
LSN    
HCS    
HuCS    
ETHR    
Agency Ties
DARPA  
NSF  
DOE  
NASA  
NIH  
NSA  
NIST  
NOAA  
EPA  
ED  
AHCPR  
VA  
Milestone Changes  
FY 1995 Actual Milestones FY 1996 Estimated Milestones FY 1997 Agency Requested Milestones
Implemented a 150 GigaFLOPS machine at ORNL running in single user mode 4Q 1994.

Upgraded machine at ORNL to multiple user/simple partition - 1Q 1995

Upgraded machine at ORNL to multiple user/overlap partition - 4Q 1995

Brought advanced architecture MP-node Intel Paragon with 1024 nodes to a state of scientific readiness.

Upgraded memory on the CM5 from 32 to 128 Gigabytes at LANL. 		Continue funding for advanced prototype systems at:
  1. Argonne National Laboratory as part of a cooperative effort to implement a standard reference message passing interface;
  2. Ames Laboratory for use in developing performance measurements and benchmarking capabilities; and
  3. Oak Ridge National Laboratory as part of the HPCRC supporting a number of the DOE Grand Challenges teams.
Coordinate HPCRP funding to optimize productivity of grand challenge application teams. Special concern to balancing CPU speed with memory, I/O, etc.