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

A question about the boundary setting
by 北尘心 01 Apr '16

01 Apr '16

Dear all, I have a question about the boundary set.In the past, I use the boundary conditions is open boundary conditions in DMRG,but now i want to eliminate the impact of open boundary,so i need to use periodic boundary conditions.In the lattices file i find its setting ,like this: <LATTICEGRAPH name = "inhomogeneous chain lattice"> <FINITELATTICE> <LATTICE ref="chain lattice"/> <EXTENT dimension="1" size ="L"/> <BOUNDARY type="periodic"/> </FINITELATTICE> <UNITCELL ref="simple1d"/> <INHOMOGENEOUS><VERTEX/></INHOMOGENEOUS> </LATTICEGRAPH> but i still have a question about the accuracy of periodic boundary conditions,because i once read an paper on this question,it points out that DMRG offers much poorer results for problems with periodic boundary conditions (PBC) than for those with open boundary conditions(OBC),.The paper is attached. I wonder whether the problem has been solved. if yes,could you give me a paper with periodic boundary conditions to verify the correctness ? In addition,if i want to set a smooth boundary condition,what should i do in the lattices file? It is highly grateful for your any suggestions. Yours sincerely, Xingbo Wei Zhejiang Normal Universtiy, Jinhua, 230014, China

2 1

A question about the boundary setting
by 北尘心 01 Apr '16

01 Apr '16

Dear all, I have a question about the boundary set.In the past, I use the boundary conditions is open boundary conditions in DMRG,but now i want to eliminate the impact of open boundary,so i need to use periodic boundary conditions.In the lattices file i find its setting ,like this: <LATTICEGRAPH name = "inhomogeneous chain lattice"> <FINITELATTICE> <LATTICE ref="chain lattice"/> <EXTENT dimension="1" size ="L"/> <BOUNDARY type="periodic"/> </FINITELATTICE> <UNITCELL ref="simple1d"/> <INHOMOGENEOUS><VERTEX/></INHOMOGENEOUS> </LATTICEGRAPH> but i still have a question about the accuracy of periodic boundary conditions,because i once read an paper on this question,it points out that DMRG offers much poorer results for problems with periodic boundary conditions (PBC) than for those with open boundary conditions(OBC),.The paper is attached. I wonder whether the problem has been solved. if yes,could you give me a paper with periodic boundary conditions to verify the correctness ? In addition,if i want to set a smooth boundary condition,what should i do in the lattices file? It is highly grateful for your any suggestions. Yours sincerely, Xingbo Wei Zhejiang Normal Universtiy, Jinhua, 230014, China

1 0

sparsediag 16-site Hubbard chain
by Chia-Chen Chang 16 Mar '16

16 Mar '16

Dear ALPS users, I ran into an issue when I tried to get the ground state energy of L=16 Hubbard 1d chain using ALPS' sparsediag. My input file is attached. The script works for L=12 and 14. With L=16, the simulation just terminated without giving any warnings or error messages. Not sure what was going on. Is it possible that my machine's running out of memory because L=16 is a rather large system in terms of Hilbert space size. Thanks for your time and help. ====================== Chia-Chen Chang cxc639(a)gmail.com

2 1

the problem of running parallelly with mps_optim
by Forwordom 14 Mar '16

14 Mar '16

Dear, M. Dolfi I have installed alps-2.2.b4 using openmpi successfully on Suse 11.0. There are not any errors during make, make test, and make install. After installation I can running successfully mps_optim serially or 1-4 CPUs. I have encountered some errors while running mps_optim parallelly using 4 or more CPUs. ...... Running task 1. Running task 1. This binary contains symmetries: nu1 This binary contains symmetries: nu1 MAQUIS repo r4143 MAQUIS repo r4143 Temporary storage enabled in /home/mlj/bin/TempFiles/n3SquareLatticeSite4.64/storage_temp_fc5a128613e7/ Temporary storage enabled in /home/mlj/bin/TempFiles/n3SquareLatticeSite4.64/storage_temp_5a9cbccacee8/ Error destroying HDF5 context of file '/home/mlj/Test/n3SquareLatticeSite4.task1.out.0.h5.tmp.4' boost::filesystem::rename: No such file or directory: "/home/mlj/Test/n3SquareLatticeSite4.task1.out.0.h5.tmp.4", "/home/mlj/mlj/Test/n3SquareLatticeSite4.task1.out.0.h5" -------------------------------------------------------------------------- mpirun noticed that process rank 3 with PID 5962 on node linux exited on signal 6 (Aborted). -------------------------------------------------------------------------- FILE ./n3SquareLatticeSite4.task1.out.h5DOES NOT EXIST! How to solve this problem? Thank you for your help. L.J. Meng Xiangtan University,Hunan Pro. China

2 1

always_measure vs ALWAYS_MEASURE and errors while measurement
by Mateusz Łącki 12 Mar '16

12 Mar '16

Dear All, I would like to compute overlap <\psi(t) | \psi(0)> for an simple open chain system using mps tools. I have encountered a series of errors while doing so. *************example 1 - input************* LATTICE='chain lattice' MODEL="spin" nsweeps=3; L=20 MAXSTATES=100 TIMESTEPS=50 DT=0.1 J=2 Jxy=0.5 measure_each=1 S_total=1/2 CONSERVED_QUANTUMNUMBERS='Sz' MEASURE[Entropy]=True COMPLEX=1; {} **************************************************** The above run with mps_optim produces an error with 2.2b4 version (lastest trunk), but with 2.2b3 it works just fine. To get it working with latest trunk I have to turn off "MEASURE[Entropy]=True" The error is: Sweep has been running for 11.71635035 seconds. Measurements. Exception raised casting True to type b WARNING: Unclosed tag: EIGENSTATES! Exception thrown: std::exception Another application I had in mind was computing the overlap. If I manage to compute the ground state (either by using 22b3 version or by turning off "MEASURE[Entropy]=True”) then I want to compute an overlap <psi(t) | psi(0)>. *************example 2 - input************* LATTICE='chain lattice' MODEL="spin" nsweeps=3; L=20 MAXSTATES=100 TIMESTEPS=50 DT=0.1 initfile='confGS.task1.out.chkp' J=2 Jxy=0.5 measure_each=1 S_total=1/2 CONSERVED_QUANTUMNUMBERS='Sz' MEASURE[Entropy]=True COMPLEX=1; MEASURE_OVERLAP[Overlap]='confGS.task1.out.chkp' ALWAYS_MEASURE=Overlap {} **************************************************** Again with 2.2b3 it works just fine, with 22b4 it produces the error: Energy -10.60920724 Measurements. Measuring Overlap overlap with confGS.task1.out.chkp. Exception raised casting True to type b Exception thrown: std::exception It works fine in 2.2.b3 alps version, while in 2.2.b4 (downloaded by svn from the "svn co https://alps.comp-phys.org/svn/alps1/trunk/alps”) it produces the following error: Measurements. Measuring Overlap overlap with confTEST.task1.out.chkp. Exception raised casting True to type b Exception thrown: std::exception Moreover it seems that in the input file for the b3 version, “always_measure=Overlap” has to be used to trigger the overlap measurements while in b4 version it is “ALWAYS_MEASURE=Overlap” Best, Mateusz Łącki

2 1

Failed tests in the make test of ALPS2.2 installation
by D´Souza, Sunil Wilfred 11 Mar '16

11 Mar '16

Dear Alps users, I have installed the ALPS2.2 package by building from the source in my laptop containing the SUSE 11.4 operating system. The cmake and make command worked fine, but then the make test command shows the following test failed 95% tests passed, 7 tests failed out of 149 The following tests FAILED: 1 - ising_single (Failed) 2 - ising (Failed) 3 - heisenberg (Failed) 4 - loop_single (Failed) 5 - exchange (Failed) 93 - matrix_unit_tests (Failed) 140 - python_pyhdf5io (Failed) Errors while running CTest make: *** [test] Error 8 after this i ran make install command and the ALPS2.2 got installed. I have run the simulation (in the python mode) on the example file "parma2a" for calculating the susceptibility using the classical Heisenberg model (spinmc binary). The result of the simulation appears to be fine. My question is that should i be concerned about the results of the make test as it is showing the failure in Ising and Heisenberg. Any suggestions to solve the problems that have poped up in the make test will be greatly appreciated. Thanks in advance for your time and consideration with best regards Sunil

2 1

Complex numbers in operators
by Simon Ydman 26 Feb '16

26 Feb '16

Dear Alps users, I want to make a spin_y operator in the spin basis. I know that spin x can be constructed in this way: <SITEOPERATOR name="Sx" site="x"> 1/2*(Splus(x)+Sminus(x)) </SITEOPERATOR> To get Sy i need to have complex i in the operator, but I am unsure of how to express it. Best regards, Simon Ydman

1 0

tutorial of maxent
by Wang Riping 23 Feb '16

23 Feb '16

Dear Developers, I am now going to use the maxent in alps for transforming data from imaginary axis to real axis. Can any one briefly tell how use this maxent or tell where I can find tutorial or examples? Best regards. Riping Wang 2016.2.23

1 0

installation error: ‘BOOST_PFTO’ has not been declared
by Wang Riping 18 Feb '16

18 Feb '16

Dear Developers, I am going to install ALPS (alps-2.2.b4). I have an error during "make". Could any one help me? [riping@fermi alps-2.2.b4.build]$ make [ 0%] Building CXX object src/alps/CMakeFiles/alps.dir/alea/observable.C.o In file included from /home/riping/software/alps-2.2.b4.src/src/alps/osiris/dump.h:249:0, from /home/riping/software/alps-2.2.b4.src/src/alps/alea/observable.h:64, from /home/riping/software/alps-2.2.b4.src/src/alps/alea/observable.C:31: /home/riping/software/alps-2.2.b4.src/src/alps/osiris/dumparchive.h:101:37: error: ‘BOOST_PFTO’ has not been declared void save_override(const T & t, BOOST_PFTO int) I use the following cmake command: cmake CMAKE_INSTALL_PREFIX=/home/riping/software/alps /home/riping/software/alps-2.2.b4.src The configuration ends normally. It can find the Boost, and I pasted the related part here. -- Looking for precompiled Boost libraries (version >= 1.52) -- Could NOT find Boost -- Boost version: 1.60.0 -- Found the following Boost libraries: -- timer -- Boost version: 1.60.0 -- Found the following Boost libraries: -- chrono -- date_time -- filesystem -- iostreams -- program_options -- python -- regex -- system -- serialization -- thread -- timer -- Adding Boost dir: /home/riping/software/boost-1.60.0/include Best regards. Riping Wang 2016.2.18

2 1

Problem in running the ALPS (spinmc) code in the parallel mode aith MPI
by D´Souza, Sunil Wilfred 15 Feb '16

15 Feb '16

Dear ALPS users, I am trying to run a spinmc simulation in the parallel mode by using the command mpiexec -np 2 spinmc --mpi parm2a.in.xml --Tmin 5 On executing the command I am getting the following error messages [LIN21:20344] *** Process received signal *** [LIN21:20344] Signal: Segmentation fault (11) [LIN21:20344] Signal code: Address not mapped (1) [LIN21:20344] Failing at address: (nil) [LIN21:20344] [ 0] /lib64/libpthread.so.0(+0xf9f0) [0x7f85f73969f0] [LIN21:20344] [ 1] /opt/openmpi/1.4.5/gcc/lib64/openmpi/mca_pml_v.so(+0x1b62) [0x7f85f1359b62] [LIN21:20344] [ 2] /opt/openmpi/1.4.5/gcc/lib64/libopen-pal.so.0(mca_base_components_close+0x76) [0x7f85f66286c6] [LIN21:20344] [ 3] /opt/openmpi/1.4.5/gcc/lib64/libmpi.so.0(mca_pml_base_select+0x301) [0x7f85f6b2cd11] [LIN21:20344] [ 4] /opt/openmpi/1.4.5/gcc/lib64/libmpi.so.0(+0x39acc) [0x7f85f6aecacc] [LIN21:20344] [ 5] /opt/openmpi/1.4.5/gcc/lib64/libmpi.so.0(MPI_Init+0x16b) [0x7f85f6b0ca8b] [LIN21:20344] [ 6] /home/sunil/ALPS2.2/lib/libalps.so.2(_ZN4alps9comm_initERiRPPcb+0x58) [0x7f85f8327088] [LIN21:20344] [ 7] /home/sunil/ALPS2.2/lib/libalps.so.2(_ZN4alps9scheduler5startEiPPcRKNS0_7FactoryE+0x37) [0x7f85f85678d7] [LIN21:20344] [ 8] spinmc(main+0x15) [0x449a85] [LIN21:20344] [ 9] /lib64/libc.so.6(__libc_start_main+0xf5) [0x7f85f454bbe5] [LIN21:20344] [10] spinmc() [0x449dd5] [LIN21:20344] *** End of error message *** -------------------------------------------------------------------------- mpiexec noticed that process rank 0 with PID 20344 on node LIN21 exited on signal 11 (Segmentation fault). -------------------------------------------------------------------------- I have tried running a band structure code in the parallel mode using the mpirun and i am having no problems in running that code. but in the case of ALPS, I am encountering the above problem. I am attaching the cmake txt file for your consideration to make sure that all the fields are correct. I have crosschecked the MPI library path in cmake txt file and i find it to be alright. I am not able to figure out what exactly is causing this problem. I am kindly requesting you to provide me help in solving this issue. I thank you in advance for your time and consideration. with best regards Sunil

2 6