May 2010 - Comp-phys-alps-users - lists.phys.ethz.ch

Problems with worms??
by Jakub Zakrzewski 27 Oct '10

27 Oct '10

Hello, Sometimes QMC worm code (used on Bose-Hubbard) terminates with the following error: here come first earlier succesful simulations then ... Created run 85 remote on Host ID: 84 Created run 86 remote on Host ID: 85 Created run 87 remote on Host ID: 86 Created run 88 remote on Host ID: 87 All processes have been assigned Checking if Simulation 1 is finished: not yet, next check in 120 seconds ( 0% done). q = -0 state1 = 0 state2 = 0 bond_type = 0 zero matrix element in remove_jump application called MPI_Abort(MPI_COMM_WORLD, -2) - process 67 Could somebody advice what I can do? Best regards Kuba

2 2

Complex numbers
by Sebastian Greschner 23 Jul '10

23 Jul '10

Hi, when I try to define a complex Hamiltonian (e.g. some extended Heisenberg-model with Sx _and_ Sy=-I/2(Splus-Sminus) couplings to external fields) fulldiag or other ALPS-programs end with the error "can not convert complex number into real one" Is this a general restriction for ALPS, or did I miss some configuration or compile option? If not - do you know any patch or work-around to enable complex-numbers? Thanks in advance and greetings, Sebastian

2 5

First ALPS 2.0 beta release
by Matthias Troyer 04 Jun '10

04 Jun '10

Dear ALPS developers and users, We have posted ALPS 2.0b1 on the web page. This is the first beta release of ALPS 2. It's main purpose is to provide a stable version to test on various platforms, and to provide feedback about installation and compilation issues. The libraries and applications should be at least as stable as version 1.3.4 was. Any potential issues should be due to dependencies on new libraries, a new build system, and support for more operating systems. Download and installation instructions are posted at http://alps.comp-phys.org/mediawiki/index.php/Download_and_install_ALPS_2 Major new features are - Use of HDF5 as binary file format for results, replacing XML for large data sets, as requested by power users. Specifying the option --write-xml with the applications will still write the old XML files in addition - Use of CMake as a new build system - Support for Windows, as requested by many users - Ready-made binary installers for MacOS X and Windows, as desired by many users. - More powerful Python based evaluation tools replacing the old XSLT based ones - Integration with the VisTrails workflow and provenance tool - A new DMFT application - Updated applications with new features - A new and more extensive set of tutorials (in progress). This first beta release will also be used to finish development on the tutorials. Progress on tutorials and documentation will be seen on the Wiki over the coming six weeks. Another beta release 2.0b2 is planned for early July, containing new features, improved documentation and tutorials and numerous improvements (and probably also bug fixes). I would like to encourage all to download and install this beta release and report problems or provide feedback on the mailing list. We want to solve all installation issued before 2.0b2 comes out. Besides all the ALPS developers I want to especially thank the following people who contributed major time to the development of new features in ALPS 2: Bela Bauer Sebastian Fuchs Lukas Gamper Jan Gukelberger Emanuel Gull Ryo Igarashi Tama Ma Olivier Parcollet Brigitte Surer Synge Todo as well as from the VisTrails team: Juliana Freire David Koop Emanuele Santos Claudio Silva For the ALPS team, Matthias Troyer

3 11

on the ALPS Pre-release 2.0b1 on windows
by jun wen 31 May '10

31 May '10

Hi All, I am a beginner to the ALPS and am eager to try the new version of ALPS. I installed the alps-2.0.0b1-windows.exe and then python and Vistrails and finally alps-vistrails-2.0.0b1-windows.exe. Then I tried to run the tutorial in the Exact Diagonalization in the directory C:\Program Files\ALPS\tutorials\ed-01-sparsediag, However, I could not either run tutorial1a.py or ed-01-sparsediag.vt. For the former, it said "ImportError: No module named pyalps", for later, it showed a command window and initializing something and reached "initializing abstraction "and would not proceed at all. I wonder if anyone could help me on this issue. Thanks a lot! Best, Jun

3 7

enegy
by "Michał Maik" 14 May '10

14 May '10

Hi, I am confused as to how the energy is calculated by the different simulations. When I use the worm or SSE algorithm I receive a different answer then when I use full diagonalization. As an example I have attached a simple parameter file below. Using worm and SSE I get the same energy while full diagonalization gives me a different answer. I have not modified this file at all, just ran it with each of the programs. I would really appreciate an explanation as to why SSE and worms do not give me the same answer as full diagonalization even at such a low temperature. Thank you. Michal LATTICE_LIBRARY="/usr/local/lib/xml/lattices.xml"; LATTICE="chain lattice"; L=6 MODEL_LIBRARY="/usr/local/lib/xml/models.xml"; MODEL="boson Hubbard"; U = 1; V = 0; mu = 0; CONSERVED_QUANTUMNUMBERS = "N" Nmax = 6; N_total=6; T=0.01 SWEEPS = 1000000 THERMALIZATION = 50000 { t = 0.08; }

2 2

IXDRDump
by Giuseppe Carleo 10 May '10

10 May '10

Hello everybody, I am currently using the ALPS (v. 1.35) scheduler in my QMC code, and everything works pretty well. Nonetheless, I've experienced an error when trying to restart my simulations on a HPC machine : "failed to read array of type double from an IXDRDump" which doesn't allow me to restart any simulation... On the other hand, on other machines the simulations are correctly restarted. I therefore assume that the way I use to restart simulations is correct, i.e. I invoke something like mpirun ./myprogram.o simulation_name.out.xml , and that the dumping of the internal variables is done correctly in my code. I think the error message should be related to some machine-specific issue. Do you have suggestions for this problem? Thank you in advance, Giuseppe

3 13

Summer school "Simulating strongly correlated systems" September 13-17
by Matthias Troyer 05 May '10

05 May '10

Dear colleagues, I would like to ask you to bring the school for which we attach a poster to the attention of any interested student or postdoc. Best regards Matthias Troyer

1 0