Re:The Future is Parallel: What's a Programmer to do? [pdf]
by baguasquirrel on Saturday June 13, 2009 @02:57AM
(#655921)
A little anecdote. A friend of mine who's working on his Ph.D at CMU sped up his physics group's simulations by an order of magnitude by simply running the same code multiple times on multiple machines. The problem was that the choice of initial random seeds could nontrivially effect the convergence time. There are examples of this in generic programming. Look at quicksort for example. Picking the right pivot can matter a great deal.
[ Parent
| on HN ]
There are no comments on this comment yet.