ALF issueshttps://git.physik.uni-wuerzburg.de/ALF/ALF/-/issues2020-11-30T12:55:39Zhttps://git.physik.uni-wuerzburg.de/ALF/ALF/-/issues/131Introduce an Op_T Container2020-11-30T12:55:39ZFlorian GothIntroduce an Op_T ContainerIn preparation for a possible OP_V container introduce an Op_T Container.
In that turn we can introduce the mscb decomposition and maybe higher order checkerboard decompositions.
In the future it seems likely that during the set up phase...In preparation for a possible OP_V container introduce an Op_T Container.
In that turn we can introduce the mscb decomposition and maybe higher order checkerboard decompositions.
In the future it seems likely that during the set up phase it might be required to have containers
that can grow. Let's test this with the Op_T array since it might turn out useful for the cb decomposition.
What we need to Do:
* [x] Introduce a C++ vector-like type for Op_T
* [ ] Figure out the interaction with the parameters of the Hamiltonian like N_Fl, N_SUN
* [x] Define a Class hierarchy so that objects conforming to an interface can be used in the container. That way we can mix dense matrix type, new sparse matrix types and the exponential types from the CB decomposition.
* [x] Check whether we can get rid of the wasteful complex matrix multiplications via that mechanism.
* [x] Implement and test a class for dense matrices. benchmark the virtualization penalty
* [ ] Include mscbdecomp https://git.physik.uni-wuerzburg.de/fgoth/mscbdecomp
* [ ] Provide an example
* [ ] Higher order checkerboard decompositions?ALF-2.0Florian GothFlorian Gothhttps://git.physik.uni-wuerzburg.de/ALF/ALF/-/issues/72improve UDV storage to handle larger scales2018-09-06T15:36:50ZFlorian Gothimprove UDV storage to handle larger scales@Hofmann has the issue that his matrices quickly exceed the capabilities of the double precision storage in D of a UDV decomposition.
Let's try whether using 2x64bit for the double resolves that issue. This should not degrade the perform...@Hofmann has the issue that his matrices quickly exceed the capabilities of the double precision storage in D of a UDV decomposition.
Let's try whether using 2x64bit for the double resolves that issue. This should not degrade the performance since all operations with D scale linearly.https://git.physik.uni-wuerzburg.de/ALF/ALF/-/issues/80Test ALF on gcc-72018-03-11T21:40:52ZFlorian GothTest ALF on gcc-7gcc-7 is out and the next major distro shipping with it will be Ubuntu 18.04 LTS. Therefore we should include that to the test system.gcc-7 is out and the next major distro shipping with it will be Ubuntu 18.04 LTS. Therefore we should include that to the test system.Florian GothFlorian Gothhttps://git.physik.uni-wuerzburg.de/ALF/ALF/-/issues/59work better with unitary matrices2017-12-01T10:58:17ZFlorian Gothwork better with unitary matricesexploit the properties of some unitary matrices.exploit the properties of some unitary matrices.Florian GothFlorian Gothhttps://git.physik.uni-wuerzburg.de/ALF/ALF/-/issues/65Filter Branch 592017-10-19T09:56:34ZFlorian GothFilter Branch 59Branch 59 contains stuff that works and stuff that doesn't work.
the simple optimizations for the Hubbard model are among the good things and the fiddling with the QR related routines
are among the things that don't make things go faster...Branch 59 contains stuff that works and stuff that doesn't work.
the simple optimizations for the Hubbard model are among the good things and the fiddling with the QR related routines
are among the things that don't make things go faster. We want to keep branch 59 for reference.Florian GothFlorian Gothhttps://git.physik.uni-wuerzburg.de/ALF/ALF/-/issues/71MPI 3 compatibility2017-10-14T18:35:16ZFlorian GothMPI 3 compatibilityWe should port our code forward to recent MPI standards.
So, current is MPI-3.1We should port our code forward to recent MPI standards.
So, current is MPI-3.1Florian GothFlorian Gothhttps://git.physik.uni-wuerzburg.de/ALF/ALF/-/issues/58Exploit the property that hopping matrices are hermitian2017-07-04T14:29:18ZFlorian GothExploit the property that hopping matrices are hermitianBetter exploit the fact that hopping matrices and their exponentials are hermitian in hop_mod.f90.Better exploit the fact that hopping matrices and their exponentials are hermitian in hop_mod.f90.Florian GothFlorian Goth