[petsc-users] Solving generalized eigenvalue problem effeciently

[Barry Smith]

> On Jun 30, 2016, at 2:00 PM, Hassan Md Mahmudulla <mhassan at miners.utep.edu> wrote:
> 
> Hi,
> I have been trying to solve Generalized eigenvalue problem where the matrix is symmetric and of size around 10,000-20,000. I have been using default solver in PETSc.

   Not enough information. Presumably you are using SLEPc? You should send email to the SLEPc mailing list for example slepc-maint at upv.es with details of how you using the solver, data from -log_summary, is the matrix sparse etc.   


> But compared to ScaLAPACK, I am having really bad performance in terms of execution time even though I am requesting only 10% eigenvalues of the spectrum. I am a new user of PETSc. So, I really don't have a clear idea on how to use it efficiently. I will appreciate if anybody can please give me an idea to use PETSc effectively so that I can get a much better performance than ScaLAPACK.
> 
> Thanks
> 
> M Hassan