Hi John, Thank you for your information. I am using the recent version of libMesh.
Also I finally figured out the culprit last weekend, which is due to hypre preconditioner used in one of my fieldsplit variables. The BoomerAMG uses very large memory during the KSPSolve() -Xujun On Mon, Jun 13, 2016 at 11:48 AM, John Peterson <jwpeter...@gmail.com> wrote: > > > On Thu, Jun 9, 2016 at 3:29 PM, Xujun Zhao <xzha...@gmail.com> wrote: > >> Derek, >> >> Excellent analysis! It really helps. >> I further looked at the PETSc log summary when running on 1 CPU, the max >> memory Petsc allocated is about 22.2G (close to what you predicted). >> However, the total memory usage is up to 100G. This is much more than >> expected. I think there should be something wrong with my API that takes >> huge memory. I will double check it. Thanks again. >> > > Was your memory spike during a solution output phase? > > This was an issue in the past, but we fixed some things (mainly avoiding > broadcasting copies of solution vectors to all procs) and it shouldn't be > as bad of a problem as it once was. > > If you are using an older version of libmesh, however, you might not have > those fixes... > > -- > John > ------------------------------------------------------------------------------ What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic patterns at an interface-level. Reveals which users, apps, and protocols are consuming the most bandwidth. Provides multi-vendor support for NetFlow, J-Flow, sFlow and other flows. Make informed decisions using capacity planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e _______________________________________________ Libmesh-users mailing list Libmesh-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/libmesh-users
