I was running 72 processes, 2 procs per node. The hostfile i use looks
like this:
master.cluster cpu=2
node1.cluster cpu=2
node2.cluster cpu=2
..............
Jeff Squyres wrote:
> I see that the alltoall2 test failed.
>
> How many processes were you running, total? Were they arranged in a
> 2-procs-per-node configuration (e.g., lamboot with all nodes marked as
> "cpu=2")?
>
>
> On Mar 7, 2005, at 8:53 PM, Ilya Lashuk wrote:
>
>> Hello everyone.
>>
>> I've istalled lam-mpi 7.1.1 on a 36 node linux cluster (Master node:
>> 2 Pentium-III processors 933MHz, 2GB memory, Tyan S2567 motherboard,
>> 35 compute nodes: 2 Pentium-III processors 933MHz, 2GB memory,
>> Supermicro 370DER motherboard, each node is running fedora core 3).
>> Most of the things go smooth, but one of the tests from
>> "lamtests-7.1.1" suite fails. Can you by a chance tell me what's
>> wrong and how it can be fixed? Please see attached files.
>>
>>
>> <files.tgz>_______________________________________________
>> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>
>
|
