IBM to Build 10 Petaflop "Mira" Super for Argonne

Mike Good, left, and Kevin Collins of IBM posed by the computer shell that will house Mira’s brains. Photo by Richard Tsong-Taatarii, Star Tribune.

Mike Good, left, and Kevin Collins of IBM posed by the computer shell that will house Mira’s brains. Photo by Richard Tsong-Taatarii, Star Tribune.

This week IBM announced a deal to build a next-generation, 10-Petaflop Blue Gene/Q supercomputer named “Mira” at the Argonne National Laboratory. Planned for deployment in 2012, the machine will have roots in Rochester, Minnesota, so the announcement garnered some special attention by Steve Alexander at the Minneapolis Star Tribune.

IBM is expected to announce Tuesday that it has received a contract from the Department of Energy to build Mira at its Rochester facility — one of the last vestiges of Minnesota’s heady supercomputer past, when Cray Research and Control Data Corp. were based in the metro area. IBM declined to reveal Mira’s purchase price, which comes out of a $180 million budget at Argonne National Laboratory in Chicago, where Mira will end up once it’s built. Rich Brueckner of Portland, Ore., who runs the InsideHPC.com (high performance computing) news website, estimates Mira will cost about $50 million, plus another $4 million to $5 million annually to pay for her electricity bills.

After talking to the reporter, I learned of a rule of thumb that says big supers cost on the order of $10 Million per Megawatt. If Mira is a similar configuration to the 14 Megawatt Blue Waters super coming to NCSA, that would put its purchase price closer to $140 Million. We’ll probably never know for sure, but maybe off quotes like this can help with your price negotiations.

Trackbacks

  1. […] This post was mentioned on Twitter by insideHPC and syoyo, Antonino Tumeo. Antonino Tumeo said: RT @insideHPC: Just in: IBM to Build 10 Petaflop "Mira" Super for Argonne http://bit.ly/g3uqdn #HPC […]

Comments

  1. blue waters cost close to $200M – at least that was what the award was from NSF – which kind of invalidates the “rule of thumb”

    • I believe that operating costs are included in that $200M figure, and just the electricity for Blue Waters will cost roughly $15M per year!