On Tue, Jul 3, 2012 at 10:50 AM, Siju George <[email protected]> wrote: > But at least you could have done a "BSDs in scientific computing" in > google before you posted this.
I did ... there are too few links. > > Just a few URLs to start with. > > http://www.freebsdnews.net/2008/04/19/freebsd-clustering/ > http://bsdnetwork.co.nr/ says nothing much. I mentioned GNU/R to mean that I am talking about highly vectorized code as opposed to more low level ways of improving code. It has built-in optimizations and so claims like 'it is inherently biased' can stand. Process scheduling by OS can be another point. Compilers ... > > In many European Universities BSD is the preferred choice for this > kind of tasks due to their stability. You got 'this kind of tasks' wrong. Anyway for the kind of stability you are talking about, one always needs to do load of tweaks. Currently GNU/Linux leads on the supercomputers too. > This was 3 years back when SMP support was maturing on dragonfly. Now > that dragonfly has much improved scalability it beats Linux in SMP > perfomance too > We will see. Best A. Mani -- A. Mani CU, ASL, CLC, AMS, CMS http://www.logicamani.in http://www.logicamani.co.cc _______________________________________________ Indian Libre User Group Cochin Mailing List http://www.ilug-cochin.org/mailing-list/ http://mail.ilug-cochin.org/mailman/listinfo/mailinglist_ilug-cochin.org #[email protected]
