Hi,
I'm using lam-6.5.x with Opterons starting from April and it works without
any problems. I should say that it works faster than mpich and successfully
runs applications that are failed under mpich.
I tested lam with 2-nodes (4CPU) test cluster and on 4-way Quartet. I'm using
gcc-3.3 to build lam. Next week we launch 16-nodes (32CPU) Opteron cluster,
so I hope to have more serious comparison between lam and mpich.
I'm also interested in running lam-7.x with gm rpi. It compiles just fine,
but test application hang in MPI_Init. The same thing happens for mpich over
gm, so I think a problem is with gm-2.0. BTW, does anybody tested lam with
gm-2.0 on other architectures?
And one more question: I'm thinking a lot of utilizing both onboard gigabit
ethernet interfaces on Opteron motherboards. I can assign them differenent
IP addresses and run MPI application as if I have separate nodes instead of
one dual-CPU node. But in such case we loose shmem communication between
MPI processes sharing this node. How hard would be patching lam to enable
binding separate MPI processes inside one nodes to different IP addresses?
Best regards,
Andrey.
--
A right thing should be simple (tm)
|
