Re: [petsc-dev] Elemental for sparse, distributed-memory, direct, quad-precision linear solves?
That depends on how many people are working on it [1] ;-) Jeff [1] https://en.wikipedia.org/wiki/The_Mythical_Man-Month On Tue, Aug 23, 2016 at 5:28 AM, Patrick Sananwrote: > Aha, cool! How much work do you estimate remains? > > On Tue, Aug 23, 2016 at 2:19 PM, Stefano Zampini > wrote: > > Interface to the sparse elemental solver is almost done in > > https://bitbucket.org/petsc/petsc/branch/stefano_zampini/ > new-elemental-interface > > > > I did not have time to finish it, and I’m not sure if Elemental now > supports > > user-defined distributions for the vectors. > > > > > > On Aug 23, 2016, at 3:15 PM, Matthew Knepley wrote: > > > > On Tue, Aug 23, 2016 at 3:01 AM, Patrick Sanan > > wrote: > >> > >> Can I use elemental to directly solve a sparse, distributed-memory > >> (MPI), quad-precision linear system? If so, with the introduction of > >> an interface, could this provide a way to support such a solve via an > >> external package for PETSc? > > > > > > I think you need Clique and I am not sure we have an interface. I would > > check first. > > > >Matt > > > > -- > > What most experimenters take for granted before they begin their > experiments > > is infinitely more interesting than any results to which their > experiments > > lead. > > -- Norbert Wiener > > > > > -- Jeff Hammond jeff.scie...@gmail.com http://jeffhammond.github.io/
Re: [petsc-dev] Elemental for sparse, distributed-memory, direct, quad-precision linear solves?
Aha, cool! How much work do you estimate remains? On Tue, Aug 23, 2016 at 2:19 PM, Stefano Zampiniwrote: > Interface to the sparse elemental solver is almost done in > https://bitbucket.org/petsc/petsc/branch/stefano_zampini/new-elemental-interface > > I did not have time to finish it, and I’m not sure if Elemental now supports > user-defined distributions for the vectors. > > > On Aug 23, 2016, at 3:15 PM, Matthew Knepley wrote: > > On Tue, Aug 23, 2016 at 3:01 AM, Patrick Sanan > wrote: >> >> Can I use elemental to directly solve a sparse, distributed-memory >> (MPI), quad-precision linear system? If so, with the introduction of >> an interface, could this provide a way to support such a solve via an >> external package for PETSc? > > > I think you need Clique and I am not sure we have an interface. I would > check first. > >Matt > > -- > What most experimenters take for granted before they begin their experiments > is infinitely more interesting than any results to which their experiments > lead. > -- Norbert Wiener > >
Re: [petsc-dev] Elemental for sparse, distributed-memory, direct, quad-precision linear solves?
It looks like Clique was merged into Elemental, so optimistically that wouldn't be an issue. On Tue, Aug 23, 2016 at 2:15 PM, Matthew Knepleywrote: > On Tue, Aug 23, 2016 at 3:01 AM, Patrick Sanan > wrote: >> >> Can I use elemental to directly solve a sparse, distributed-memory >> (MPI), quad-precision linear system? If so, with the introduction of >> an interface, could this provide a way to support such a solve via an >> external package for PETSc? > > > I think you need Clique and I am not sure we have an interface. I would > check first. > >Matt > > -- > What most experimenters take for granted before they begin their experiments > is infinitely more interesting than any results to which their experiments > lead. > -- Norbert Wiener
Re: [petsc-dev] Elemental for sparse, distributed-memory, direct, quad-precision linear solves?
Interface to the sparse elemental solver is almost done in https://bitbucket.org/petsc/petsc/branch/stefano_zampini/new-elemental-interface I did not have time to finish it, and I’m not sure if Elemental now supports user-defined distributions for the vectors. On Aug 23, 2016, at 3:15 PM, Matthew Knepleywrote: > On Tue, Aug 23, 2016 at 3:01 AM, Patrick Sanan > wrote: > Can I use elemental to directly solve a sparse, distributed-memory > (MPI), quad-precision linear system? If so, with the introduction of > an interface, could this provide a way to support such a solve via an > external package for PETSc? > > I think you need Clique and I am not sure we have an interface. I would check > first. > >Matt > > -- > What most experimenters take for granted before they begin their experiments > is infinitely more interesting than any results to which their experiments > lead. > -- Norbert Wiener
Re: [petsc-dev] Elemental for sparse, distributed-memory, direct, quad-precision linear solves?
On Tue, Aug 23, 2016 at 3:01 AM, Patrick Sananwrote: > Can I use elemental to directly solve a sparse, distributed-memory > (MPI), quad-precision linear system? If so, with the introduction of > an interface, could this provide a way to support such a solve via an > external package for PETSc? > I think you need Clique and I am not sure we have an interface. I would check first. Matt -- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener
[petsc-dev] Elemental for sparse, distributed-memory, direct, quad-precision linear solves?
Can I use elemental to directly solve a sparse, distributed-memory (MPI), quad-precision linear system? If so, with the introduction of an interface, could this provide a way to support such a solve via an external package for PETSc?