By Matthew Dublin
It looks like Big Blue has pulled the plug on the "Blue Waters" project, a Power7-based petaflop class supercomputer that it was contracted to development for the National Center for Supercomputing Applications (NCSA) at the University of Illinois.
Both the NCSA and IBM have jointly stated that the project was killed because "The innovative technology that IBM ultimately developed was more complex and required significantly increased financial and technical support by IBM beyond its original expectations. NCSA and IBM worked closely on various proposals to retain IBM's participation in the project but could not come to a mutually agreed-on plan concerning the path forward."
Certainly not a good sign for exascale computing if this petascale project proved too much for IBM and the NCSA...
The project, which kicked off in 2007, had the goal of building a sustained-petaflop supercomputer, which obviously required considerable innovation on the part of IBM in terms of not only processing capability, but more importantly, power efficiency. The electric bills that would be generated from a petaflop supercomputer being run at capacity or at near capacity even during just "regular" business hours would be astronomical at best.
The NCSA has stated that it hasn't given up hope on the delivery of a feasible petaflop supercomputer yet. The site has already tapped the National Science Foundation to ensure that the NCSA's efforts aren't in vain and that a Blue Waters-like system could still be delivered.