Quoting LAI Loong Fong <lailf@bii.a-star.edu.sg>: > Hi Chris, > I tried sending you and the Bioclusters mailing list the following email on > the 20th March, wondering did you receive it. > > Regards > LAI Loong-Fong > > On the topic on low latency interconnects, how many of you are running your > clusters with such interconnets? Especially in the area of bioinformatics. > We are using quadrics on one IA32 cluster and Myrinet on IA32, IA64 and G5. > With occasional mpi jobs that uses these interconnects, I do not see other > jobs requiring it. We certainly don't use any low latency interconnects here at the Sanger. We always considered it more cost effective just to use the money to buy more nodes. As you say, there's relatively little current bioinformatics software that requires MPI; many of our problems are embarrassingly parallel and can be broken into small enough independent subproblems that they can be run on poorly interconnected small machines. The only thing a decent interconnect would gain us is a better position on the Top 500 list. :-) Tim