<HTML> <div> <div>Hi Juerg,</div> <div>thank you very much for the help. I submitted a test calculation fro H2 and it works. Now I will examine different basis sets for my system.</div> <div> </div> <div>best regards,</div> <div>Petko </div> </div> <div>================</div> <div align="left">Dr. Petko St. Petkov Faculty of Chemistry University of Sofia 1126 Sofia, Bulgaria tel: +359 2 8161 433 </div> <div>=====================</div> <blockquote style="padding-right: 0px; padding-left: 5px; margin-left: 5px; border-left: 2px solid rgb(0, 0, 0); margin-right: 0px;"> -----Original Message----- From: Juerg Hutter <h...@pci.uzh.ch> To: cp...@googlegroups.com Date: Fri, 24 Sep 2010 13:15:05 +0200 (MEST) Subject: Re: [CP2K:2855] problems using HSE as a hybrid functional. <div style="font-family: monospace,courier new,courier;">Hi you have most likely a memory problem. But in order to be sure you should run a test calculation for a small system (e.g. H2) with a small basis set. Your input takes on my machine about 450 MB per core. Another point is that the MOLOPT basis sets are not well suited for HFX calculations. Because of their difuse primitives you will have very large CPU times. You have to use either the new ADMM method or another type of basis. Using the DZV-shortrange basis from BASIS_SET, the integral calculation takes about 1 minute and the SCF iterations about 2 seconds on my 8 core machine (ca. 250 MB memory/core). best regards Juerg ---------------------------------------------------------- Juerg Hutter Phone : ++41 44 635 4491 Physical Chemistry Institute FAX : ++41 44 635 6838 University of Zurich E-mail: hut...@pci.uzh.ch Winterthurerstrasse 190 CH-8057 Zurich, Switzerland ---------------------------------------------------------- On Thu, 23 Sep 2010, Petko Petkov wrote: > Hi CP2K users and developers, > > > > I need a hybrid functional for my research. I have submitted a test job for > > > > > Pt8 cluster using HSE functional. I took the &XC section from the > > cp2k/tests/QS/regtest-hybrid/NE-hybrid-HSE06-lda.inp file. The input file is > > > > > > attached. The job was submitted on IBM BG/P using 128 cores in SMP mode. > > After that I got a error like: > > > > > > > > " HFX_MEM_INFO| Est. max. program size before HFX [MB's]: > > > 0 > > > > * *** 09:45:01 ERRORL2 in > > hfx_energy_potential:integrate_four_center *** > > > > *** processor 0 err=-300 condition > FAILED at line 413 *** * > > > > > > ===== Routine Calling Stack ===== > > 11 integrate_four_center > > > > 10 qs_ks_build_kohn_sham_matrix > > > > 9 qs_ks_update_qs_env > > > > 8 init_scf_loop > > > > 7 scf_env_do_scf > > > > 6 qs_energies_scf > > > > 5 qs_forces > > > > 4 cp_eval_at > > > > 3 geoopt_bfgs > > > > 2 cp_geo_opt > > > > 1 CP2K > > > > CP2K| condition FAILED at line 413 > > > > CP2K| Abnormal program termination, stopped by process number > > 0....... > > > > ................................................................................................................................................." > > > > I have no idea if this error comes from the code or from the wrong > > input. If I remove the &HF section the job is working well. I checked > > > the google group and found the same settings for the &XC section. The > > > code was recompiled from our sys administrator with libint, necessary for > > > HFX. The output is attached. > > > > Thanks a lot in advance! > > > > > > Best Regards, > > Petko > > ================ > > Dr. Petko St. Petkov > > Faculty of Chemistry > > University of Sofia > > 1126 Sofia, Bulgaria > > tel: +359 2 8161 433 > > ===================== > > -- > You received this message because you are subscribed to the Google Groups "cp2k" group. > To post to this group, send email to cp...@googlegroups.com. > To unsubscribe from this group, send email to cp2k+uns...@googlegroups.com. > For more options, visit this group at <a target="_blank" href="http://groups.google.com/group/cp2k?hl=en"> http://groups.google.com/group/cp2k?hl=en</a>. > > -- You received this message because you are subscribed to the Google Groups "cp2k" group. To post to this group, send email to cp...@googlegroups.com. To unsubscribe from this group, send email to cp2k+uns...@googlegroups.com. For more options, visit this group at <a target="_blank" href="http://groups.google.com/group/cp2k?hl=en"> http://groups.google.com/group/cp2k?hl=en</a>. </div> </blockquote> </HTML>