Page 23 - GCN, Jan/Feb 2016

“Basically what happens is the asset is first offered to any other federal agency to see if they want it,” Broughton said. “The second priority is to offer it to state and local govern- ments. And then, from there, it is offered out to the public.”
In theory, that’s a bargain for the prospective new own- ers, but they have to come and get the equipment. And that means figuring out how to take it out of its original space and reinstall it however and wherever the new owners like.
“If someone actually wanted the system to be workable, they would need Cray to come in and take it down and take it apart and reinstall in a professional manner,” he added, and the cost of such repurposing can range as high as $400,000.
Finally, if there aren’t any takers via GSAXcess, it’s usually off to the wood chipper.
A HAPPY RETIREMENT
For a lucky few, however, a program called PRObE — for Par- allel Reconfigurable Observational Environment — seeks to repurpose supercomputers for researchers and students who otherwise might not have access to such systems. The pro- gram is a partnership of Los Alamos National Laboratory, the National Science Foundation (which provides the funding),
Carnegie Mellon University, the University of Utah and the University of New Mexico.
Supercomputers “usually get put in a truck and [sent] to a big chipper and ground down to dust,” said Andree Ja- cobson, CIO at the New Mexico Consortium, which runs the PRObE program. “And there are some people in government who think that isn’t a good thing because it is often a waste of money because the computers are very powerful.”
And although those hand-me-down systems might not be the most up-to-date models, they can still provide valuable research opportunities.
However, the process is not simple or quick because the en- tire supercomputer does not get a happy retirement.
“We can never receive any storage — hard drives, flash memories, any kinds of those things where data has been stored,” he said. “That is considered a security risk. But the computers themselves — the processors, memories, mother- board —are not used to store that kind of data. So with the right person who advocates for it, there are ways the hard- ware could be repurposed.”
It would be easier if one could just leave the system where it is, Jacobson added, but that isn’t what usually happens,
wide variety of challenges, from trying to follow supernova patterns to designing Pringles potato chips that won’t fly off the conveyor belt.
Taking Hopper’s place at NERSC is Cori, named after Gerty Cori, the first American woman to win a Nobel Prize
in science. Months ago, Cray provided
a smaller version of Cori so users could start transferring their applications. Now Cori is going through the final testing stages before it formally replaces Hopper. Bashor said NERSC always has “two sys- tems on the floor,” so while Cori is going through acceptance testing, the Edison supercomputer system is handling the production load that Hopper formerly managed.
Broughton said NERSC usually has a small party in honor of each supercom- puter’s retirement. This year, however, the agency skipped the celebration because employees were busy moving
to a new facility on Lawrence Berkeley National Laboratory’s campus. Grace Hopper wasn’t one to fuss or get hung
up on formalities — she reportedly didn’t bat an eye when she was named the Data
Processing Management Association’s Computer Sciences Man of the Year in 1969 — so she probably wouldn’t mind.
— Suzette Lohmeyer
GCN JANUARY/FEBRUARY 2016 • GCN.COM 19
FLICKR.COM/DEPARTMENTOFENERGY