What caught my eye was a configurable pool of heterogenous nodes, each with massive (to me) amounts of local storage and well-adapted to matrix operations. I got a picture in my mind's eye of J-like primitives linked in a "tacit" program.
The NVIDIA contribution was hazy in the Livescience article. A better clue is given here: https://en.wikipedia.org/wiki/Summit_(supercomputer) . I guess the IBM Summit isn't just a bunch of graphics cards. :-) On Sat, Jun 16, 2018 at 12:32 PM, Raul Miller <[email protected]> wrote: > If they're using graphics cards as the base, it's not so much like J's > arrays, but more a collection of Nx4 matrices. > > Thanks, > > -- > Raul > On Sat, Jun 16, 2018 at 7:29 AM Ian Clark <[email protected]> wrote: > > > > Yet more ways of coming up with "42" in time for yesterday. > > > > https://www.livescience.com/62827-fastest-supercomputer.html > > > > If you read the whole article you'll see they could have had a computer > > with the J primitives as its machine instructions. > > ---------------------------------------------------------------------- > > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
