From creaman1986 at gmail.com Wed Apr 3 15:09:41 2013 From: creaman1986 at gmail.com (=?Big5?B?rFK8/bOv?=) Date: Wed, 3 Apr 2013 21:09:41 +0800 Subject: [Wannier] occupation of Wannier function Message-ID: Hi guys! I have a question about Wannier function. Can we get the occupation number of wannier function through this Wannier90 code? thanks! -------------- next part -------------- An HTML attachment was scrubbed... URL: From ivo_souza at ehu.es Wed Apr 3 19:49:42 2013 From: ivo_souza at ehu.es (Ivo Souza) Date: Wed, 3 Apr 2013 17:49:42 +0000 (UTC) Subject: [Wannier] occupation of Wannier function In-Reply-To: References: Message-ID: > Can we get the occupation number of wannier function through this Wannier90 code? At present the code does not provide that information. I can think of a couple of ways in which this could be done: (i) Calculate the DOS projected onto the desired WF, and integrate up to the Fermi level. The calculation of the orbital-projected DOS will be included in the next release of wannier90. (ii) Calculate the BZ average of the occupation matrix in the Wannier gauge [see Eqs. 16 and 23 of PRB 85, 014435 (2012) for the definition]. The orbital occupations are given by the diagonal elements. Best, Ivo From a.mostofi at imperial.ac.uk Wed Apr 3 19:54:52 2013 From: a.mostofi at imperial.ac.uk (Arash Mostofi) Date: Wed, 03 Apr 2013 18:54:52 +0100 Subject: [Wannier] occupation of Wannier function In-Reply-To: References: Message-ID: <515C6CEC.4080407@imperial.ac.uk> Hello there, Adding to Ivo's reply: Without disentanglement the occupation numbers are always 1 (or 2 in the case of spin degeneracy). With disentanglement, and just considering the Gamma-point of the BZ for simplicity, the occupation "number" becomes an occupation matrix f_{ij}. See, for example, the discussion leading up to Eq (9) for f_{ij} in Andrinopoulos et al, J Chem Phys 135, 154105 (2011); http://link.aip.org/link/doi/10.1063/1.3647912 Computing f_{ij} is not implemented in v1.2 of Wannier90, but will be in the next release. It is fairly straightforward to code and would look something like this: ! aam: calculate V = U_opt . U do s=1,num_wann do k=1,ndim do m=1,num_wann v_matrix(k,s) = v_matrix(k,s) + u_matrix_opt(k,m,1)*u_matrix(m,s,1) enddo enddo enddo ! aam: calculate f = V^dagger . V do r=1,num_wann do s=1,num_wann do k=1,ndim f_w(r,s) = f_w(r,s) + v_matrix(k,s) * conjg(v_matrix(k,r)) enddo enddo enddo Best wishes, Arash -- Dr Arash A Mostofi Departments of Materials and Physics Deputy Director, CDT on Theory and Simulation of Materials Imperial College London, London SW7 2AZ, UK +44 (0)207 594 8154 | www.cmth.ph.ic.ac.uk/people/a.mostofi On 03/04/2013 14:09, ??? wrote: > Hi guys! > > I have a question about Wannier function. > > Can we get the occupation number of wannier function through this > Wannier90 code? > > thanks! > > > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier > From nicolas.lorente at cin2.es Thu Apr 4 08:15:19 2013 From: nicolas.lorente at cin2.es (Nicolas Lorente) Date: Thu, 04 Apr 2013 08:15:19 +0200 Subject: [Wannier] occupation of Wannier function In-Reply-To: References: Message-ID: <515D1A77.7050401@cin2.es> On 03/04/13 19:49, Ivo Souza wrote: >> Can we get the occupation number of wannier function through this Wannier90 code? > At present the code does not provide that information. I can think of a > couple of ways in which this could be done: > > (i) Calculate the DOS projected onto the desired WF, and integrate up > to the Fermi level. The calculation of the orbital-projected DOS will be > included in the next release of wannier90. > > (ii) Calculate the BZ average of the occupation matrix in the Wannier > gauge [see Eqs. 16 and 23 of PRB 85, 014435 (2012) for the definition]. > The orbital occupations are given by the diagonal elements. > > Best, > > Ivo > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier To follow on idea (i), I sent our wpdos.f90 code to compute the PDOS using wannier90 in this thread: http://www.democritos.it/pipermail/wannier/2012-November/000603.html Best, Nicol?s -- Nicol?s Lorente CIN2 (CSIC) CIN2 building, UAB 08193 Bellaterra, Spain Tel. +34-937373622 From creaman1986 at gmail.com Thu Apr 4 11:11:56 2013 From: creaman1986 at gmail.com (=?Big5?B?rFK8/bOv?=) Date: Thu, 4 Apr 2013 17:11:56 +0800 Subject: [Wannier] Wannier Digest, Vol 63, Issue 1 In-Reply-To: References: Message-ID: Thanks for all of your replies they are really helpful! 2013/4/4 > Send Wannier mailing list submissions to > wannier at quantum-espresso.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.democritos.it/mailman/listinfo/wannier > or, via email, send a message with subject or body 'help' to > wannier-request at quantum-espresso.org > > You can reach the person managing the list at > wannier-owner at quantum-espresso.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Wannier digest..." > > > Today's Topics: > > 1. occupation of Wannier function (???) > 2. Re: occupation of Wannier function (Ivo Souza) > 3. Re: occupation of Wannier function (Arash Mostofi) > 4. Re: occupation of Wannier function (Nicolas Lorente) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Wed, 3 Apr 2013 21:09:41 +0800 > From: ??? > Subject: [Wannier] occupation of Wannier function > To: wannier at quantum-espresso.org > Message-ID: > < > CAE35B+8Y04xWB3S7_Q2WSKWnqDN-Lg1xEo8x7hRVG6gSd6ZX2A at mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > Hi guys! > > I have a question about Wannier function. > > Can we get the occupation number of wannier function through this Wannier90 > code? > > thanks! > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > http://www.democritos.it/pipermail/wannier/attachments/20130403/e3476e2c/attachment.html > > ------------------------------ > > Message: 2 > Date: Wed, 3 Apr 2013 17:49:42 +0000 (UTC) > From: Ivo Souza > Subject: Re: [Wannier] occupation of Wannier function > To: =?GB2312?B?6sXosOqQ?= > Cc: wannier at quantum-espresso.org > Message-ID: > Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed > > > Can we get the occupation number of wannier function through this > Wannier90 code? > > At present the code does not provide that information. I can think of a > couple of ways in which this could be done: > > (i) Calculate the DOS projected onto the desired WF, and integrate up > to the Fermi level. The calculation of the orbital-projected DOS will be > included in the next release of wannier90. > > (ii) Calculate the BZ average of the occupation matrix in the Wannier > gauge [see Eqs. 16 and 23 of PRB 85, 014435 (2012) for the definition]. > The orbital occupations are given by the diagonal elements. > > Best, > > Ivo > > > ------------------------------ > > Message: 3 > Date: Wed, 03 Apr 2013 18:54:52 +0100 > From: Arash Mostofi > Subject: Re: [Wannier] occupation of Wannier function > To: wannier at quantum-espresso.org > Message-ID: <515C6CEC.4080407 at imperial.ac.uk> > Content-Type: text/plain; charset=UTF-8; format=flowed > > Hello there, > > Adding to Ivo's reply: > > Without disentanglement the occupation numbers are always 1 (or 2 in the > case of spin degeneracy). With disentanglement, and just considering the > Gamma-point of the BZ for simplicity, the occupation "number" becomes an > occupation matrix f_{ij}. See, for example, the discussion leading up to > Eq (9) for f_{ij} in > > Andrinopoulos et al, J Chem Phys 135, 154105 (2011); > http://link.aip.org/link/doi/10.1063/1.3647912 > > Computing f_{ij} is not implemented in v1.2 of Wannier90, but will be in > the next release. It is fairly straightforward to code and would look > something like this: > > ! aam: calculate V = U_opt . U > do s=1,num_wann > do k=1,ndim > do m=1,num_wann > v_matrix(k,s) = v_matrix(k,s) + u_matrix_opt(k,m,1)*u_matrix(m,s,1) > enddo > enddo > enddo > > ! aam: calculate f = V^dagger . V > do r=1,num_wann > do s=1,num_wann > do k=1,ndim > f_w(r,s) = f_w(r,s) + v_matrix(k,s) * conjg(v_matrix(k,r)) > enddo > enddo > enddo > > Best wishes, > > Arash > > -- > Dr Arash A Mostofi > Departments of Materials and Physics > Deputy Director, CDT on Theory and Simulation of Materials > Imperial College London, London SW7 2AZ, UK > +44 (0)207 594 8154 | www.cmth.ph.ic.ac.uk/people/a.mostofi > > On 03/04/2013 14:09, ??? wrote: > > Hi guys! > > > > I have a question about Wannier function. > > > > Can we get the occupation number of wannier function through this > > Wannier90 code? > > > > thanks! > > > > > > > > > > _______________________________________________ > > Wannier mailing list > > Wannier at quantum-espresso.org > > http://www.democritos.it/mailman/listinfo/wannier > > > > > ------------------------------ > > Message: 4 > Date: Thu, 04 Apr 2013 08:15:19 +0200 > From: Nicolas Lorente > Subject: Re: [Wannier] occupation of Wannier function > To: wannier at quantum-espresso.org > Message-ID: <515D1A77.7050401 at cin2.es> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > On 03/04/13 19:49, Ivo Souza wrote: > >> Can we get the occupation number of wannier function through this > Wannier90 code? > > At present the code does not provide that information. I can think of a > > couple of ways in which this could be done: > > > > (i) Calculate the DOS projected onto the desired WF, and integrate up > > to the Fermi level. The calculation of the orbital-projected DOS will be > > included in the next release of wannier90. > > > > (ii) Calculate the BZ average of the occupation matrix in the Wannier > > gauge [see Eqs. 16 and 23 of PRB 85, 014435 (2012) for the definition]. > > The orbital occupations are given by the diagonal elements. > > > > Best, > > > > Ivo > > _______________________________________________ > > Wannier mailing list > > Wannier at quantum-espresso.org > > http://www.democritos.it/mailman/listinfo/wannier > > To follow on idea (i), I sent our wpdos.f90 code to compute the PDOS > using wannier90 in this thread: > > http://www.democritos.it/pipermail/wannier/2012-November/000603.html > > Best, > > Nicol?s > > > > -- > Nicol?s Lorente > CIN2 (CSIC) > CIN2 building, UAB > 08193 Bellaterra, Spain > Tel. +34-937373622 > > > > ------------------------------ > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier > > > End of Wannier Digest, Vol 63, Issue 1 > ************************************** > -------------- next part -------------- An HTML attachment was scrubbed... URL: From creaman1986 at gmail.com Sun Apr 7 09:32:17 2013 From: creaman1986 at gmail.com (=?Big5?B?rFK8/bOv?=) Date: Sun, 7 Apr 2013 15:32:17 +0800 Subject: [Wannier] supercell calculation Message-ID: Hi there I am trying to calculate 5 d-like Wannier function centered on Cobalt atom in a nanojunction system with 82 Cu atoms and 1 Co atom in the middle of the junction,and I compare my results between 3x3x3 K-points sampling and gamma point sampling. I found Wannier function that come from gamma point sampling has much d-wave character than 3x3x3 K-points sampling and the later one looks like a mess in the unit supercell. By the way,I use the disentangle procedure and my outer window is determined by including all high peaks that calculated by PWSCF-PDOS calculation ,and there are about 50~60 state per K-point. is there anything wrong with my procedure? thanks! -------------- next part -------------- An HTML attachment was scrubbed... URL: From creaman1986 at gmail.com Mon Apr 8 02:28:12 2013 From: creaman1986 at gmail.com (=?Big5?B?rFK8/bOv?=) Date: Mon, 8 Apr 2013 08:28:12 +0800 Subject: [Wannier] Wannier Digest, Vol 63, Issue 3 In-Reply-To: References: Message-ID: Hi there Sorry,I forgot to post my input file. ------------------------------------------------------------------------------------------------- num_wann = 5 dis_num_iter = 5000 num_iter = 300 num_bands = 552 write_xyz = true wannier_plot = true wannier_plot_format = xcrysden wannier_plot_list = 1 2 3 4 5 wannier_plot_supercell = 1 dis_win_min = 6.2 dis_win_max = 8.5 hr_plot = true spin = up iprint = 5 begin unit_cell_cart Ang 9.852329297 0.019428883 0.000000000 -4.909330837 8.543756106 0.000000000 0.0000000000 0.000000000 16.46176642 end_unit_cell_cart Begin Projections Co:d End Projections begin atoms_frac Cu 0.499473878 0.50554763 0.192178394 Cu 0.752366094 0.512987566 0.185087084 Cu 0.998427849 0.508716899 0.18519496 Cu 0.247814593 0.507977268 0.184759921 Cu 0.497252911 0.757349328 0.185225584 Cu 0.747262849 0.759415304 0.186521036 Cu 0.998399403 0.759685457 0.184952543 Cu 0.247262136 0.757726017 0.184159785 Cu 0.496432791 0.006737385 0.185418141 Cu 0.747040256 0.009322876 0.184434411 Cu 0.996494146 0.008724441 0.184195288 Cu 0.245453219 0.006703916 0.184554256 Cu 0.492357829 0.252987551 0.1848503 Cu 0.74460048 0.260499653 0.190459673 Cu 0.996016795 0.258222464 0.184236469 Cu 0.245714611 0.257866164 0.185487153 Cu 0.414382421 0.590649326 0.063379766 Cu 0.662866053 0.591594647 0.063236567 Cu 0.9134024 0.59171742 0.061679198 Cu 0.163540917 0.591204469 0.061178705 Cu 0.413803218 0.841543761 0.061608038 Cu 0.66334598 0.841695853 0.061970218 Cu 0.912829839 0.841291891 0.061972862 Cu 0.163321593 0.84147119 0.061395077 Cu 0.413487161 0.091737396 0.061597889 Cu 0.663694106 0.09316416 0.062620131 Cu 0.91315335 0.091845438 0.060855559 Cu 0.163698512 0.092046257 0.061492197 Cu 0.413767944 0.342480981 0.062846678 Cu 0.663325883 0.341955675 0.061955209 Cu 0.911971068 0.341476191 0.062521948 Cu 0.163099809 0.341591866 0.061519669 Cu 0.330171487 0.425521751 0.938053004 Cu 0.579642576 0.425661225 0.938176 Cu 0.8299247 0.425309443 0.937805914 Cu 0.079222982 0.42473908 0.937575561 Cu 0.330450683 0.674813779 0.937834548 Cu 0.579746102 0.675079573 0.938128642 Cu 0.829401298 0.675090428 0.938057943 Cu 0.079786405 0.675195682 0.937801606 Cu 0.33002575 0.925354428 0.937922529 Cu 0.580373753 0.925875501 0.937821538 Cu 0.829680793 0.925026142 0.937847185 Cu 0.079756363 0.925109235 0.937839413 Cu 0.330120406 0.17570481 0.937907035 Cu 0.579955554 0.175300161 0.937769458 Cu 0.829531064 0.175536253 0.937481576 Cu 0.080130079 0.175492614 0.937639532 Cu 0.414351398 0.590707502 0.812760752 Cu 0.662883575 0.591458244 0.813439549 Cu 0.913466443 0.591739076 0.814077251 Cu 0.163632078 0.591395349 0.814216003 Cu 0.413815261 0.841747328 0.814089116 Cu 0.663287708 0.841793609 0.813966083 Cu 0.912994338 0.841328595 0.813880993 Cu 0.163454496 0.841509329 0.814260055 Cu 0.413348926 0.091597959 0.814060762 Cu 0.663616144 0.093148591 0.812601397 Cu 0.913223613 0.091799834 0.814302104 Cu 0.163786332 0.091934408 0.814080861 Cu 0.413692344 0.34231516 0.813303605 Cu 0.66325113 0.341841855 0.813896566 Cu 0.911955794 0.341513234 0.812528085 Cu 0.163231977 0.341688473 0.813895853 Cu 0.49941011 0.505724946 0.684656405 Cu 0.75208049 0.512918048 0.691002732 Cu 0.998420007 0.508872864 0.690092811 Cu 0.24789872 0.508065504 0.690839828 Cu 0.497153555 0.757354263 0.690798897 Cu 0.747130516 0.759224478 0.690033575 Cu 0.998413406 0.759640287 0.690872657 Cu 0.247344441 0.757904561 0.691306346 Cu 0.496393045 0.006844671 0.68999176 Cu 0.746974797 0.009186295 0.690867632 Cu 0.996604963 0.008631914 0.691243372 Cu 0.24549993 0.006720824 0.690957851 Cu 0.492504693 0.253213377 0.691028529 Cu 0.744563442 0.260622245 0.684600945 Cu 0.996116213 0.258225365 0.690861255 Cu 0.245910172 0.257951663 0.690200318 Cu 0.619371564 0.385340078 0.301255388 Cu 0.620275633 0.38493744 0.574524509 Co 0.622392238 0.382793659 0.437849026 end atoms_frac mp_grid : 3 3 3 begin kpoints 0.0000000 0.0000000 0.0000000 0.0000000 0.0000000 0.3333333 0.0000000 0.0000000 -0.3333333 0.0000000 0.3333333 0.0000000 0.0000000 0.3333333 0.3333333 0.0000000 0.3333333 -0.3333333 0.0000000 -0.3333333 0.0000000 0.0000000 -0.3333333 0.3333333 0.0000000 -0.3333333 -0.3333333 0.3333333 0.0000000 0.0000000 0.3333333 0.0000000 0.3333333 0.3333333 0.0000000 -0.3333333 0.3333333 0.3333333 0.0000000 0.3333333 0.3333333 0.3333333 0.3333333 0.3333333 -0.3333333 0.3333333 -0.3333333 0.0000000 0.3333333 -0.3333333 0.3333333 0.3333333 -0.3333333 -0.3333333 -0.3333333 0.0000000 0.0000000 -0.3333333 0.0000000 0.3333333 -0.3333333 0.0000000 -0.3333333 -0.3333333 0.3333333 0.0000000 -0.3333333 0.3333333 0.3333333 -0.3333333 0.3333333 -0.3333333 -0.3333333 -0.3333333 0.0000000 -0.3333333 -0.3333333 0.3333333 -0.3333333 -0.3333333 -0.3333333 end kpoints ------------------------------------------------------------------------------------------------- 2013/4/7 > Send Wannier mailing list submissions to > wannier at quantum-espresso.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.democritos.it/mailman/listinfo/wannier > or, via email, send a message with subject or body 'help' to > wannier-request at quantum-espresso.org > > You can reach the person managing the list at > wannier-owner at quantum-espresso.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Wannier digest..." > > > Today's Topics: > > 1. supercell calculation (???) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Sun, 7 Apr 2013 15:32:17 +0800 > From: ??? > Subject: [Wannier] supercell calculation > To: wannier at quantum-espresso.org > Message-ID: > Lgjemqf-Gtemfw492VNrwtijTyGJQ at mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > Hi there > > I am trying to calculate 5 d-like Wannier function centered on Cobalt > atom in a nanojunction system with 82 Cu atoms and 1 Co atom in the middle > of the junction,and I compare my results between 3x3x3 K-points sampling > and gamma point sampling. > > I found Wannier function that come from gamma point sampling has much > d-wave character than 3x3x3 K-points sampling and the later one looks like > a mess in the unit supercell. > > By the way,I use the disentangle procedure and my outer window is > determined by including all high peaks that calculated by PWSCF-PDOS > calculation ,and there are about 50~60 state per K-point. > > is there anything wrong with my procedure? > > thanks! > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > http://www.democritos.it/pipermail/wannier/attachments/20130407/8b8f649e/attachment.html > > ------------------------------ > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier > > > End of Wannier Digest, Vol 63, Issue 3 > ************************************** > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nickpapior at gmail.com Mon Apr 8 11:47:00 2013 From: nickpapior at gmail.com (Nick Papior Andersen) Date: Mon, 8 Apr 2013 11:47:00 +0200 Subject: [Wannier] Plans for parallel version? Message-ID: Dear, I have tried searching for whether it is envisioned to do a parallel version (not just library openmp parallelization) of the code in the archive, to no avail. So, are there any plans of implementing an MPI, MPI/OpenMP hybrid or full OpenMP, parallel version of the program? Kind regards Nick -------------- next part -------------- An HTML attachment was scrubbed... URL: From jonathan.yates at materials.ox.ac.uk Mon Apr 8 13:08:33 2013 From: jonathan.yates at materials.ox.ac.uk (Jonathan Yates) Date: Mon, 8 Apr 2013 11:08:33 +0000 Subject: [Wannier] Plans for parallel version? In-Reply-To: References: Message-ID: <78877D222077A642A4CFAB09BD0077E0CC175F@MBX10.ad.oak.ox.ac.uk> On 8 Apr 2013, at 10:47, Nick Papior Andersen wrote: > > Dear, > > I have tried searching for whether it is envisioned to do a parallel version (not just library openmp parallelization) of the code in the archive, to no avail. > > So, are there any plans of implementing an MPI, MPI/OpenMP hybrid or full OpenMP, parallel version of the program? Dear Nick, The 2.0 version of Wannier90 is in its final stages before release - this will offer MPI parallelism for the calculation of properties ie distributing energies or kpoints for spectral properties. However, we have not attempted to parallelise the core disentanglement and wannierisation routines. Personally, I've always felt that some sort of shared memory parallelisation would make more sense than a full distributed memory MPI implementation. Even for a big systems the memory requirements are rather small and there is not an obvious way to distribute the data. It would be an interesting problem to address. The main reason we haven't looked at this more seriously is simply lack of time. If you, or anyone else, have ideas about this, Arash and I would be happy to discuss further. Yours Jonathan -- Department of Materials, University of Oxford, Parks Road, Oxford, OX1 3PH, UK tel: +44 (0)1865 612797 http://users.ox.ac.uk/~oums0549/ From nickpapior at gmail.com Tue Apr 9 00:35:52 2013 From: nickpapior at gmail.com (Nick Papior Andersen) Date: Mon, 8 Apr 2013 23:35:52 +0100 Subject: [Wannier] Plans for parallel version? In-Reply-To: <78877D222077A642A4CFAB09BD0077E0CC175F@MBX10.ad.oak.ox.ac.uk> References: <78877D222077A642A4CFAB09BD0077E0CC175F@MBX10.ad.oak.ox.ac.uk> Message-ID: Dear Jonathan, I have only recently used the wannier program, so have not really delved into the code structure.?I was merely interested in the development plans and progress made. On a note of parallalism. My typical experiences with parallel programming is the following: It seems *almost* obvious to utilize a shared memory model for almost any program with large matrices (whence I think this would also be applicable for the Wannier90 code), and ok, this is a large statement, but small changes often leads to better shared parallelisation. It typically requires limited effort and will yield a nice speed up. MPI is *much* more difficult to get a nice scalability with, and often a restructure of the serial program is required to obtain a reasonable speed up. Hybrids (are for me) the best way to go. MPI on the *bigger* picture, OpenMP on the *local* picture. It seems to me that the disentanglement is quite an entangled computation :). In which case I fully agree with you, the OpenMP should be the way to go. Do you have a public site for following the development version? Or is that off limits to the public? However, it is nice to know that you have the parallelisation under consideration (albeit the time issue). I look forward to see your implementation details of Wannier90 in parallel. Kind regards Nick 2013/4/8 Jonathan Yates > > > On 8 Apr 2013, at 10:47, Nick Papior Andersen wrote: > > > > > Dear, > > > > I have tried searching for whether it is envisioned to do a parallel version (not just library openmp parallelization) of the code in the archive, to no avail. > > > > So, are there any plans of implementing an MPI, MPI/OpenMP hybrid or full OpenMP, parallel version of the program? > > Dear Nick, > > ?The 2.0 version of Wannier90 is in its final stages before release - this will offer MPI parallelism for the calculation of properties ie distributing energies or kpoints for spectral properties. However, we have not attempted to parallelise the core disentanglement and wannierisation routines. > > ?Personally, I've always felt that some sort of shared memory parallelisation would make more sense than a full distributed memory MPI implementation. Even for a big systems the memory requirements are rather small and there is not an obvious way to distribute the data. It would be an interesting problem to address. The main reason we haven't looked at this more seriously is simply lack of time. If you, or anyone else, have ideas about this, Arash and I would be happy to discuss further. > > ?Yours > ? Jonathan > > > > > > -- > Department of Materials, University of Oxford, Parks Road, Oxford, OX1 3PH, UK > tel: +44 (0)1865 612797 ? ? ? ? ? ? ? ?http://users.ox.ac.uk/~oums0549/ > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier From haoweipeng at gmail.com Thu Apr 18 20:26:43 2013 From: haoweipeng at gmail.com (Haowei Peng) Date: Thu, 18 Apr 2013 12:26:43 -0600 Subject: [Wannier] When will the new version available? Message-ID: As the title, Thanks a lot. -- Haowei Peng National Renewable Energy Laboratory 15013 Denver West Pkwy Golden, CO 80401 -------------- next part -------------- An HTML attachment was scrubbed... URL: From haoweipeng at gmail.com Fri Apr 19 20:06:07 2013 From: haoweipeng at gmail.com (Haowei Peng) Date: Fri, 19 Apr 2013 12:06:07 -0600 Subject: [Wannier] test Message-ID: att -- Haowei Peng National Renewable Energy Laboratory 15013 Denver West Pkwy Golden, CO 80401 -------------- next part -------------- An HTML attachment was scrubbed... URL: From lcyin at imr.ac.cn Wed Apr 24 05:26:48 2013 From: lcyin at imr.ac.cn (=?GBK?B?0vzA+7Ok?=) Date: Wed, 24 Apr 2013 11:26:48 +0800 (GMT+08:00) Subject: [Wannier] How to set projection block in .win file Message-ID: <10abdc2.d1f3.13e3a14490e.Coremail.lcyin@imr.ac.cn> Dear colleague, I am a beginner of wannier90 code. I want to know, are there any thumb of rules to set the projection block in .win file for a specific system? Let us say CaB6. I set "random" for this system, it seems the calculation is very huge. Then, I set B: l=0; l=1; Ca: l=0; l=2; Then wannier90 output file says, too much defined projections. How to set the projection block, by which we can quickly get the MLWF results. Gratitude for any suggestion. -- ======================================== Li-Chang Yin, Ph.D. ------------------ Advanced Carbon Division Shenyang Materials Science National Laboratory Institute of Metal Research Chinese Academy of Sciences 72 Wenhua Road,Shenyang 110016, China +86-24-83978238 (office) +86-13889378291 (mobile) +86-24-23891320 (FAX) E-mail: lcyin at imr.ac.cn ======================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: From julen.azpiroz at gmail.com Wed Apr 24 09:24:36 2013 From: julen.azpiroz at gmail.com (Julen Ibanez Azpiroz) Date: Wed, 24 Apr 2013 09:24:36 +0200 Subject: [Wannier] How to set projection block in .win file In-Reply-To: <10abdc2.d1f3.13e3a14490e.Coremail.lcyin@imr.ac.cn> References: <10abdc2.d1f3.13e3a14490e.Coremail.lcyin@imr.ac.cn> Message-ID: Dear Li-Chang, The number of projections you choose must be in accordance with the number of Wannier functions you consider. In your example, you have chosen 4 orbitals for B (one s and three p) and 6 orbitals for Ca (one s and 5 d), so totally I think you have chosen 10 orbitals. If you get the error message 'too much defined projections', surely you have chosen less Wannier functions than 10 (what is your value for input parameters num_wann or num_bands)? Hope this helps, best regards, Julen On Wed, Apr 24, 2013 at 5:26 AM, ??? wrote: > Dear colleague, > > I am a beginner of wannier90 code. > I want to know, are there any thumb of rules to set the projection block > in .win file for a specific system? > Let us say CaB6. I set "random" for this system, it seems the calculation > is very huge. Then, I set > > B: l=0; l=1; > Ca: l=0; l=2; > Then wannier90 output file says, too much defined projections. > How to set the projection block, by which we can quickly get the MLWF > results. > Gratitude for any suggestion. > > > > > -- > ======================================== > Li-Chang Yin, Ph.D. > ------------------ > Advanced Carbon Division > Shenyang Materials Science National Laboratory > Institute of Metal Research > Chinese Academy of Sciences > 72 Wenhua Road,Shenyang 110016, China > > +86-24-83978238 (office) > +86-13889378291 (mobile) > +86-24-23891320 (FAX) > E-mail: lcyin at imr.ac.cn > ======================================== > > > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier > > -------------- next part -------------- An HTML attachment was scrubbed... URL: