The new ultra super computer planned by IBM and the US Government will be the most powerful computer ever created, containing over one and a half million processors and clocking up performance speeds of up to 20 petaflops.
The Sequoia system will surpass the previous contender for the fastest computer by far, the IBM Roadrunner in Los Almos, which has a processing speed of just over 1 petaflops and is one of only 2 computers in the world capable of Petaflops speeds. The other computer that broke the Petaflops barrier was developed by the Japanese.
Ok if you don’t understand exactly what a petaflops is you are not alone, neither do most of us. In the simplest of terms, flops stands for floating point operations per second and is a measure of a computer’s processing speed and how many mathematical calculations it can do in one second but in reality it’s a bit more complex than that. One petaflops amounts to a thousand trillion floating operations per second.
To get an idea of how powerful the Sequoia supercomputer is likely to be, IBS reckon its power will amount to more than the combined systems of every single machine in the top 500 supercomputer list. Wow!
The Sequoia system is forecasted for delivery in early 2012 and will be constructed at IBM’s BlueGene facilities in Minnesota before being housed in the Lawrence Livermore National Laboratory (LLNL) in California on 96 racks.
Initial plans are to deploy it for maintaining nuclear weapons and simulating nuclear tests but it could also be used for other purposes including making accurate weather predictions.
Ok so the sequoia might be the latest in the race to create the most powerful and the fastest computer but the rate at which technology is expanding no doubt it won’t be too long before we’re hearing about computer speeds in exaflops, zettaflops or yottaflops, the question is, who will get there first?
If you want to find out all the latest news on tech why not subscribe to our RSS feed?
Well…
Last couple of years ago many of our technical guy knows about the development of a gigaflops supercomputer. Even I myself own one cluster server that clock a gigaflops.
After that then come a teraflops which now has easily be taken by a petaflops. Even my small cluster server running under OpenMosix (10.56 gigaflops) eat much electricity and cooling system. For sure this new supercomputer could gives some green house effect for their power.
I’m planning to start my teraflops cluster server using Beowulf cluster technology in the middle of 2009 (with personal financing).
Not nearly as much if its an organic processor