[Beowulf] Rocks Cluster and 2 Ethernet networks
mechti01 at luther.edu
Sat Oct 30 21:54:41 PDT 2004
Thanks for your help. I have not installed Rocks just yet. Can you
explain to me what the ch_p4 device of MPI_CH is? Nochmals, vielen Dank!
>> I'm considering installing the Rocks cluster distro on a cluster that
>> only ethernet. As I understand it, eth0 (or first network interface) is
>> used for administration and also message passing if no other high speed
>> interface is present (e.g. myrinet). My question is, if each of my
>> compute node have two ethernet interfaces, say eth0 and eth1, can the
>> cluster be configured that message passing takes place only over eth1?
>> would be nice to have an interface devoted to just message passing. If
>> is possible, how would I go about setting it up? If it's not possible,
>> there are a lot of performance loss due to the fact that other tasks
>> as administration, etc.) are also taking place over eth0? Thanks in
>> advance for your help.
> do you want to use the ch_p4 device of MPICH for communication? Then you
> have to set the machinefile for mpirun to include only the names of the
> interface in all nodes. Maybe your queuingsystem can do this already for
> Furthermore, you have to change the setting in mpirun.args that way, that
> will be substituded with the name of the second interface. E.g.
> MPI_HOST=`hostname | sed "s/^node/internal/"`
> to change the name from node001 to internal001 or whatever names you use.
> Otherwise your machinefile will be scanned in a wrong way (wrong
> of the processes to the nodes in the end), and the communication back from
> slaves to the head node of the job will still use the wrong interface. You
> simply include this at the beginning of the mpirun.arg file. If it's
> set, it will no be set later in the script.
> Cheers - Reuti
More information about the Beowulf