Monday, October 03, 2011
Intel starts with government financed exaflops project !
The new company will initially have offices in Oregon, California and the Washington, D.C and will expand to other government ventures soon.
BAD HARDWARE WEEK:
H.R. act 2354
Latest Title: Energy and Water Development and Related Agencies Appropriations Act, 2012
Sponsor: Rep Frelinghuysen, Rodney P. [NJ-11] (introduced 6/24/2011 House approved 7/15/2011) says:
Not enough emergency computing capacities?
For example, in 2011, the Los Alamos lab projects it will have about 6.5 times the capacity in teraflops than Livermore and nearly 19 times the capacity of Sandia. As a result, there will be a significant centralization of supercomputing capacity in Los Alamos. Were Los Alamos supercomputers to suffer damage, neither Livermore nor Sandia would obviously possess sufficient capacity to assume the workload.
How to aggregate efforts when Intel isn't aggregatee its own? Only Skymont architecture should show enough energy efficiency needed for exaflops desgin. Together with MIC multicore.
Exaflops supercomputer should solve all the missing teraflops lab problems :)
The new supercomputer won't be ready until 2018 though, with the lengthy wait attributed to the need to reinvent entire facets of computing. The problem here has to do with fundamental computing limitations; an exaflop-capable supercomputer will need to shift a prodigious amount of data, but the efficiency required to hit these speeds simply does not exist yet.
BAD HARDWARE WEEK:
H.R. act 2354
Latest Title: Energy and Water Development and Related Agencies Appropriations Act, 2012
Sponsor: Rep Frelinghuysen, Rodney P. [NJ-11] (introduced 6/24/2011 House approved 7/15/2011) says:
Department of Energy "has not yet aggregated exascale research components into a coherent effort."
Not enough emergency computing capacities?
For example, in 2011, the Los Alamos lab projects it will have about 6.5 times the capacity in teraflops than Livermore and nearly 19 times the capacity of Sandia. As a result, there will be a significant centralization of supercomputing capacity in Los Alamos. Were Los Alamos supercomputers to suffer damage, neither Livermore nor Sandia would obviously possess sufficient capacity to assume the workload.
How to aggregate efforts when Intel isn't aggregatee its own? Only Skymont architecture should show enough energy efficiency needed for exaflops desgin. Together with MIC multicore.
Exaflops supercomputer should solve all the missing teraflops lab problems :)
The new supercomputer won't be ready until 2018 though, with the lengthy wait attributed to the need to reinvent entire facets of computing. The problem here has to do with fundamental computing limitations; an exaflop-capable supercomputer will need to shift a prodigious amount of data, but the efficiency required to hit these speeds simply does not exist yet.
THIS LINK posted by Joseph Bell : 3:19 PM
