Actually the new intels are pretty similar to the AMDs these days. There are 2 things here: - multiple-cores per chip - and multiple chips
For eg: one can buy 2-chip-dual-core = 4CPU machine. [with AMD each chip has a separate memory bank. With intel, there is a single controller with multiple banks. But when 1 chip is used - only half the memory banks are accessed - or something like that] So in both AMD and Intel, when both chips [each chip - a dual-core] are used, MB available scales up - as compared to 1 chip usage. However within a chip [i.e dual core] - the MB from main memory to cpu/cache is same irrespective of both cores being used or only one. So when both are used - the effective memory bandwith is not scaling up. So to get best parallel speedup - one should choose 'np' as 'no_of_memory banks' - not 'no_of_cpus'. So, on this 2x2 = 4CPU machine, I suspect the best performance scaling can be seen only with '-np 2' Wrt MPICH on SMP, we were sugested to use the following MPICH configure options: --with-pm=gforker --device=ch3:nemesis --enable-fast Satish On Tue, 13 Nov 2007, Barry Smith wrote: > > It depends on how the memory is connected to the individual cores or CPUS; > for example the AMD has a different approach than Intel. If the different > processors/cores > have SEPERATE paths to memory then you will not see this terrible effect. > > Barry > > > > On Nov 13, 2007, at 10:23 AM, Gideon Simpson wrote: > > > This is also true for a multi-processor machine, or its unique to multi-core > > machines? > > -gideon > > > > On Nov 13, 2007, at 11:14 AM, Barry Smith wrote: > > > > > > > > Not possible. The problem is that with one process it uses all the memory > > > bandwidth, when you change to use 2 processes (2 cores) each core > > > now gets only half the memory bandwidth and hence essentially half > > > the speed. > > > > > > Barry > > > > > > > > > Barry > > > > > > On Nov 13, 2007, at 10:06 AM, Gideon Simpson wrote: > > > > > > > Has anyone had any success in getting good performance on multi-core > > > > intel os x machines with petsc? What's the right way to get MPICH up > > > > and running for such a thing? > > > > > > > > -Gideon Simpson > > > > Department of Applied Physics and Applied Mathematics > > > > Columbia University > > > > > > > > > > > > > >
