For the former case, you are not trying to squeeze one subdomain to one compute node, are you? Why is it memory limited?
----- Original Message ----- From: "Jed Brown" <[email protected]> To: "For users of the development version of PETSc" <petsc-dev at mcs.anl.gov> Sent: Friday, March 1, 2013 11:53:35 PM Subject: Re: [petsc-dev] Deflated Krylov solvers for PETSc Obviously that depends on how fast you want the answer, but anywhere from significantly less than one thousand (limited by memory capacity) to about 1M (limited by latency and strong scalability).
