Hi All, Thanks for the suggestions, I will look into dsyevx. Unfortunately cuSolver might be out of the question since I actually need to compute these eigenvalues within a device kernel, although it might be worthwhile to compute the eigenvalues with cusolver as a pre-processing step to my own kernel.
Best regards, Jacob Faibussowitsch (Jacob Fai - booss - oh - vitch) > On Jul 7, 2021, at 14:13, Adam Denchfield <[email protected]> wrote: > > syevjBatched from cuSolver is quite good once it's configured fine. It's a > direct solve for all eigenpairs, works for batches of small matrices with > sizes up to (I believe) 32x32. The default CUDA example using it works except > if you have "too many" small matrices, in which case you'll overload the GPU > memory and need to further batch the calls. I found it to be fast enough for > my needs. > > Regards, > Adam Denchfield > Ph.D Student, Physics > University of Illinois in Chicago > B.S. Applied Physics (2018) > Illinois Institute of Technology > Email: [email protected] <mailto:[email protected]> > > > On Wed, Jul 7, 2021 at 2:31 AM Jose E. Roman <[email protected] > <mailto:[email protected]>> wrote: > cuSolver has syevjBatched, which seems to fit your purpose. But I have never > used it. > > Lanczos is not competitive for such small matrices. > > Jose > > > > El 6 jul 2021, a las 21:56, Jed Brown <[email protected] > > <mailto:[email protected]>> escribió: > > > > Have you tried just calling LAPACK directly? (You could try dsyevx to see > > if there's something to gain by computing less than all the eigenvalues.) > > I'm not aware of a batched interface at this time, but that's what you'd > > want for performance. > > > > Jacob Faibussowitsch <[email protected] <mailto:[email protected]>> > > writes: > > > >> Hello PETSc/SLEPc users, > >> > >> Similar to a recent question I am looking for an algorithm to compute the > >> smallest eigenvalue and eigenvector for a bunch of matrices however I have > >> a few extra “restrictions”. All matrices have the following properties: > >> > >> - All matrices are the same size > >> - All matrices are small (perhaps no larger than 12x12) > >> - All matrices are SPD > >> - I only need the smallest eigenpair > >> > >> So far my best bet seems to be Lanczos but I’m wondering if there is some > >> wunder method I’ve overlooked. > >> > >> Best regards, > >> > >> Jacob Faibussowitsch > >> (Jacob Fai - booss - oh - vitch) >
