[R-sig-hpc] Rmpi and node swamping

[Thomas Hampton]

I am using qsusb, Rmpi and papply.

Anyone out there using this combo?

I am running into a problem where all my R slaves end up on the
first two nodes, swamping them.

Not even sure how you would do this in you intended to, because
I am not actually specifying which processes go where.

Anyway, any insight you all might have would be greatly appreciated.

Thanks,

Tom