Date: Thu, 15 May 1997 18:07:21 -0400 (EDT) From: Ben Black <black@zen.cypher.net> To: "Pedro F. Giffuni" <pgiffuni@fps.biblos.unal.edu.co> Cc: Jonathan Lemon <jlemon@americantv.com>, hackers@FreeBSD.ORG Subject: Re: Cluster Computing in BSD Message-ID: <Pine.LNX.3.91.970515180626.10796C-100000@zen.cypher.net> In-Reply-To: <Pine.A41.3.95.970515161320.38526A-100000@fps.biblos.unal.edu.co>
next in thread | previous in thread | raw e-mail | index | archive | help
the SP2 is probably a bad example. MCI here just replaced theirs with a pair of Ultra Enterprise 6000s and performance is great. the moral is: one parallel computer does not benchmark for them all. b3n On Thu, 15 May 1997, Pedro F. Giffuni wrote: > On Thu, 15 May 1997, Jonathan Lemon wrote: > > > On May 05, 1997 at 01:55:53PM -0700, Pedro F. Giffuni wrote: > > > Parallel computing is not cost effective, > > > > Hmm. Show references please. You might want to check out a paper that > > discusses this issue: > > > It's my personal conclusion with the systems we have on campus. We have an > SP2, with 4 processors (which in theory is the best price/performance you > can get) and the guys that handle it report it isn't fast and that they > need more resources for it. > I must admit that even with all the courses they have received from IBM > they have showed to be very incompetent in different areas, but with all > those IBM engineers coming in and out (my campus is very important in a > national level) for about two years now, I would expect more results. > Maybe parallel computing is cost effective but I haven't noticed it :(. > > > D.Wood and M.Hill, "Cost-Effective Parallel Computing", > > IEEE Computer, Feb 1995 > > > Thanks for the reference...I'll look for it. > > Pedro. > > > -- > > Jonathan > > >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.3.91.970515180626.10796C-100000>