January 2009 - Comp-phys-alps-users

alps in ubuntu hardy
by sinan bulut 17 Apr '09

17 Apr '09

Does anyone know or have experience on how to install alps in ubuntu(hardy 8.04) ? I tried to install it from the source using "configure, make ... etc" commands but it did not work.

4 17

Extending a simulation
by Jakub Zakrzewski 29 Jan '09

29 Jan '09

Hello, I am fresh, just trying ALPS, so question may be trivial. How to "extend a simulation" in QMC (worm). Info in wiki: http://alps.comp-phys.org/mediawiki-1.9.3/index.php/Running says to increase the amount of sweeps (which I did) and set a simulation status ti running from finished. But where to set it? Example -- ... is missing... Help appreciated KubaZakrzewski

4 6

Re: [ALPS-users] extending a simulation
by Jakub Zakrzewski 29 Jan '09

29 Jan '09

Thanks to all who answered. Synge - your practical advice most appreciated. I do not know whether it applies to all MC-codes. I run QMC (worm) on different twists of Bose-Hubbard. Mathias. Thanks. That I found after asking. BUT still- ... you wrote > Date: Wed, 28 Jan 2009 09:31:00 -0800 > From: Matthias Troyer <troyer(a)phys.ethz.ch> > > If your parameters are called parms, then parameter2xml creates a file > > parms.in.xml, and the simulation creates a file parms.out.xml . The > latter is the file to edit. > I did exactly as you suggest - My parameter file is called prl I changed status from "finished" to "running" in prl.out.xml. I increased SWEEPS in prl.task1.out.xml. Then I execute worm --Tmin 50 parms.out.xml and get info: - I copy the screen here kuba@b52:~/Pr3 $ worm --Tmin 50 prl.out.xml Worm algorithm quantum Monte Carlo simulation v2.0 copyright (c) 1997-2007 by Simon Trebst <trebst(a)comp-phys.org> and Matthias Troyer <troyer(a)comp-phys.org> for details see the publication: A.F. Albuquerque et al., J. of Magn. and Magn. Materials 310, 1187 (2007). using the ALPS parallelizing scheduler copyright (c) 1994-2006 by Matthias Troyer <troyer(a)comp-phys.org>. see Lecture Notes in Computer Science, Vol. 1505, p. 191 (1998). based on the ALPS libraries version 1.3.3 available from http://alps.comp-phys.org/ copyright (c) 1994-2008 by the ALPS collaboration. Consult the web page for license details. For details see the publication: A.F. Albuquerque et al., J. of Magn. and Magn. Materials 310, 1187 (2007). parsing task files ... Loading information about run 1 from file /users/champ/kuba/Pr3/prl.task1.out.run1 Cannot change parameter SWEEPS Cannot open simulation file /users/champ/kuba/Pr3/prl.task1.out.xml. Task 1 does not exist. Finished with everything. Please note the info Cannot change parameter SWEEPS. So what I am doing wrongly? regards Kuba (Jakub Zakrzewski)

1 0

Problem in Running the Program
by zhian asadzadeh 27 Jan '09

27 Jan '09

Dear All I installed the ALPS package successfully but when I tried to run the program gave following error: *error while loading shared libraries: libmkl_lapack.so: cannot open shared object file: No such file or directory* I have lapack library on my system, too. Thanks Zhian

2 1

Re: [ALPS-users] one solution to compilation issues with GCC (Jeff Hammond)
by Tadeusz Wasiutynski 27 Jan '09

27 Jan '09

Thank you Jeff, now it works fine. I did not face above problem with older compiler. Tadeusz On Sun, 25 Jan 2009 12:00:03 +0100, comp-phys-alps-users-request wrote > > What do you mean break? GCC 4.3.1 segfaulted on the compilation of > spinmc_factory.C: > > ============================================== > g++ -DHAVE_CONFIG_H -I. -I. -I../.. -DNDEBUG > -I/software/alps/include -I/software/alps/boost_1_37_0 > -I/software/alps/include -I/software/alps/lp_solve_4.0 -pthread > -ftemplate-depth-150 -O3 -DBOOST_DISABLE_ASSERTS -fexceptions -w -MT > spinmc_factory.o -MD -MP -MF .deps/spinmc_factory.Tpo -c -o > spinmc_factory.o spinmc_factory.C > spinmc_factory.C: In constructor > 'AbstractSpinSim<MAT>::AbstractSpinSim(const alps::ProcessList&, > const alps::Parameters&, int) [with MAT = DiagMatrix<double, 3>]': > spinmc_factory.C:194: internal compiler error: Segmentation fault > Please submit a full bug report, with preprocessed source if appropriate. > See <http://bugs.opensuse.org/> for instructions. > make: *** [spinmc_factory.o] Error 1 > ============================================== > > but it compiled when I changed the "-O3" in CXXFLAGS "-O0" (both "- > O2" and "-O1" failed as well for the same reasons) I was able to compile > that file. I then stopped the build after the spinmc_factory.o was > formed, restored optimization to the default, and the rest of ALPS > compilation went fine. > > I should have mentioned this the last time but I wasn't sure if it > was just by distro that had this issue. I have 4 GB so the segfault > is certainly a bug not the absence of sufficient memory. > > I have not reported this bug to GCC because such things usually > require posting the code that fails and I have no hope of doing this > properly without including all of ALPS/Boost. It would be great it > someone at ALPS who is closer to the GCC project could do so though. > > Jeff > > On Sat, Jan 24, 2009 at 4:52 AM, Tadeusz Wasiutynski > <Tadeusz.Wasiutynski(a)ifj.edu.pl> wrote: > > > > > > Indeed the compilation of ALPS-1.3.3 goes smoothly with gcc-4.3.1 and > > boost-1.37. Unfortunately alps-applications compilation breaks on > > spinmc.factory while was not the case with older gcc. All other prerequisities > > but lp_solve come in the system. Any suggestions? > > Shall we compare all the components incl kernel? > > Tadeusz > > > > > > -- > Jeff Hammond > The University of Chicago > http://home.uchicago.edu/~jhammond/ > > End of Comp-phys-alps-users Digest, Vol 34, Issue 9 > ***************************************************

4 7

2D Heisenberg model with open boundary in one spatial direction
by Fu-Jiun Jiang ITP 26 Jan '09

26 Jan '09

Hello all I think I should try to make my quick questions more precise (hence slightly more interesting). For a 2D Heisenberg model with open boundary in one spatial direction, can one measure the spin-spin correlations as well as the on-site magnitization as functions of distance from the open boundary using ALPS without modifying the codes dramatically? Thanks in advance! Sincerely, FJ

4 8

follow-up on GCC 4.3 bug for spinmc_factory
by Jeff Hammond 26 Jan '09

26 Jan '09

I discovered that the problem with g++ 4.3 segfaulting on spinmc_factory has been logged with GCC Bugzilla - http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38369 - and that the problem persists in 4.3.3, which I discovered on my own prior to finding that bug report. According to the bug report, GCC 4.4.0 corrects whatever the problem is, so ALPS spinmc users reliant upon GCC might want to skip from GCC 4.2 to 4.4 (whenever it appears) or resign themselves to the poor performance of "-O0" with 4.3 as reported previously. Jeff -- Jeff Hammond The University of Chicago http://home.uchicago.edu/~jhammond/

2 1

Re: [ALPS-users] one solution to compilation issues with GCC
by Tadeusz Wasiutynski 24 Jan '09

24 Jan '09

On Fri, 23 Jan 2009 12:00:04 +0100, comp-phys-alps-users-request wrote > 8. Re: one solution to compilation issues with GCC (Matthias > Troyer) > > ---------------------------------------------------------------------- > > > Message: 7 > Date: Thu, 22 Jan 2009 10:59:55 -0600 > From: Jeff Hammond <jeff.science(a)gmail.com> > Subject: Re: [ALPS-users] one solution to compilation issues with GCC > To: comp-phys-alps-users(a)phys.ethz.ch > Message-ID: > <96f4bb620901220859g1561bd29uc938624e70e5e502(a)mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1 > > I had no trouble compiling ALPS 1.33 with GCC 4.3.1 and Boost 1.37, > and yes, it was much faster to do this than to install GCC. > > Jeff > > On Thu, Jan 22, 2009 at 9:36 AM, Matthias Troyer > <troyer(a)phys.ethz.ch> wrote: > > > > Thanks for your help! It might be easier to switch Boost versions than > > compiler. Could you or someone else with the same problem try to download > > Boost 1.37.0 and see whether that also solves the problem? > > > > Matthias > > > > -- > Jeff Hammond > The University of Chicago > http://home.uchicago.edu/~jhammond/ > > ------------------------------ > > Message: 8 > Date: Thu, 22 Jan 2009 10:53:04 -0800 > From: Matthias Troyer <troyer(a)phys.ethz.ch> > Subject: Re: [ALPS-users] one solution to compilation issues with GCC > To: comp-phys-alps-users(a)phys.ethz.ch > Message-ID: <BF5A683A-B460-4E06-9E27-67260C11C400(a)phys.ethz.ch> > Content-Type: text/plain; charset=US-ASCII; format=flowed > > On 22 Jan 2009, at 08:59, Jeff Hammond wrote: > > > I had no trouble compiling ALPS 1.33 with GCC 4.3.1 and Boost 1.37, > > and yes, it was much faster to do this than to install GCC. > > > > Jeff > > Thanks, I'll add a note to the Wiki! > > Matthias > > End of Comp-phys-alps-users Digest, Vol 34, Issue 7 > *************************************************** Indeed the compilation of ALPS-1.3.3 goes smoothly with gcc-4.3.1 and boost-1.37. Unfortunately alps-applications compilation breaks on spinmc.factory while was not the case with older gcc. All other prerequisities but lp_solve come in the system. Any suggestions? Shall we compare all the components incl kernel? Tadeusz

2 1

Re: [ALPS-users] uniform susceptibility
by Tadeusz Wasiutynski 24 Jan '09

24 Jan '09

On Sat, 24 Jan 2009 12:00:02 +0100, comp-phys-alps-users-request wrote > 4. Re: uniform susceptibility (Matthias Troyer) > Message: 4 > Date: Fri, 23 Jan 2009 09:21:06 -0800 > From: Matthias Troyer <troyer(a)phys.ethz.ch> > Subject: Re: [ALPS-users] uniform susceptibility > To: comp-phys-alps-users(a)phys.ethz.ch > Message-ID: <71F122F5-D527-4E3A-B308-8A86A23FD70C(a)phys.ethz.ch> > Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes > > On 23 Jan 2009, at 08:58, Tadeusz Wasiutynski wrote: > >> > > Thank you for prompt answer. There is still something unclear for > > me. When > > running spinmc Ising model with say L=8 and T=1, square lattice, > > update > > cluster, sweeps 100000 I get: > > Magnetization^2 0.99862084 > > |Magnetization| 0.999285325 > > Susceptibility 63.911734 > > So it looks like Susceptibility=N * Magnetization^2 > > On the other hand given susceptibility at Tc scales very well with > > size. > > Where do I make mistake? > > The "susceptibility" is measuredas N*Magnetizatio^2/T. Since T=1 > here, you get what you observe > > Matthias > > ------------------------------ > In the other words it is not (<Magnetization^2>-<|Magnetization|>^2)/T as one could think. tadeusz

1 0

Re: [ALPS-users] uniform susceptibility
by Tadeusz Wasiutynski 23 Jan '09

23 Jan '09

On Fri, 23 Jan 2009 12:00:04 +0100, comp-phys-alps-users-request wrote > 2. uniform susceptibility (Tadeusz Wasiutynski) > Message: 2 > Date: Thu, 22 Jan 2009 12:55:44 +0100 > From: "Tadeusz Wasiutynski" <Tadeusz.Wasiutynski(a)ifj.edu.pl> > Subject: [ALPS-users] uniform susceptibility > To: comp-phys-alps-users(a)phys.ethz.ch > Message-ID: <20090122115137.M49749(a)ifj.edu.pl> > Content-Type: text/plain; charset=iso-8859-2 > > hello, > I try to understand all examples in alps-applications. How can I > figure out what exactly is in the output as "uniform susceptibility" > (say in Ising) and why it is not simply <M2>-<M>2? In which part of > the code it is hidden? Tadeusz > > ------------------------------ > > Message: 4 > Date: Thu, 22 Jan 2009 23:10:11 +0900 (JST) > From: Synge Todo <wistaria(a)comp-phys.org> > Subject: Re: [ALPS-users] uniform susceptibility > To: comp-phys-alps-users(a)phys.ethz.ch, Tadeusz.Wasiutynski(a)ifj.edu.pl > Message-ID: <20090122.231011.245963879.wistaria(a)comp-phys.org> > Content-Type: Text/Plain; charset=us-ascii > > If 'M' means the total uniform magnetization (i.e. M = \sum s_i^z), > the uniform susceptibility and <M2>-<M>2 differ by the factor \beta/N > (\beta = 1/T and N the number of sites). If 'M' is the magnetization > per site (\sum s_i^z / N), then the factor becomes \beta \times N. > > Synge > Thank you for prompt answer. There is still something unclear for me. When running spinmc Ising model with say L=8 and T=1, square lattice, update cluster, sweeps 100000 I get: Magnetization^2 0.99862084 |Magnetization| 0.999285325 Susceptibility 63.911734 So it looks like Susceptibility=N * Magnetization^2 On the other hand given susceptibility at Tc scales very well with size. Where do I make mistake? Tadeusz

2 1