Thank you for your kindly reply! On Sat, Jul 24, 2010 at 9:28 AM, Backlund, Daniel <[email protected]>wrote:
> > When you restart with UseSaveData, SIESTA will check to see if a DM file is > present. This DM file is updated > after each self-consistent step. If found, SIESTA will check if it has the > correct dimensions and is compatible > with the the FDF used as input. Then SIESTA will continue with the DM. If > your job is a CG run, then SIESTA > will inquire the CG file and use it for the next CG step. > > Daniel J. Backlund > Texas Tech University - Physics > [email protected] > ________________________________________ > From: Bin Shao [[email protected]] > Sent: Friday, July 23, 2010 8:06 PM > To: [email protected] > Subject: Re: [SIESTA-L] How to stop and continue the calculation? > > I still have something puzzled me. If I set UseSaveData to true, the siesta > will continue the calculation or just restart it by using the Data from the > last calculation. > > On Fri, Jul 23, 2010 at 10:10 AM, Bin Shao <[email protected]<mailto: > [email protected]>> wrote: > Thank you for your kindly reply! > > I will follow your suggestion and try it later. > > > On Fri, Jul 23, 2010 at 9:36 AM, Backlund, Daniel <[email protected] > <mailto:[email protected]>> wrote: > Just a note for the google purposes. Be sure you do not modify your FDF > file in such a way > the dimensions of the DM file will change (MeshCutoff, lattice vectors, > basis set, etc.). I do not > think it would be too difficult to implement a STOP procedure. You could > modify siesta.F (in 2.0.2) > and add some proper code like this: > > character(len=label_length+5) :: filstop > logical :: foundstop > > filstop = ' ' > filstop = paste( slabel, '.STOP' ) > > inquire (file=fname, exist=foundstop) > > if(foundstop .and. IOnode) call die("Found STOP file. Exiting") > > Of course, you would need decide whether the stop check is done after each > SCF or MD step and > then put the code in the proper place. There are probably more lines needed > to make it pretty. > > You would recompile, and then touch an slabel.STOP file in the run > directory. > > Daniel J. Backlund > Texas Tech University - Physics > [email protected]<mailto:[email protected]> > ________________________________________ > From: Bin Shao [[email protected]<mailto:[email protected]>] > Sent: Thursday, July 22, 2010 7:49 PM > To: [email protected]<mailto:[email protected]> > Subject: Re: [SIESTA-L] How to stop and continue the calculation? > > Thank you for your reply. > > I recheck the manual ant notice that flag DM.UseSaveDM to use the DM from > the last calculation. > > It seems that siesta has not the function to stop the calcuation after the > next interation by creating a file like .stop, or before the interation, the > program will check whether there is file like .stop, if not, the calculation > will continue. > > 2010/7/23 Marcos Veríssimo Alves <[email protected]<mailto: > [email protected]><mailto:[email protected] > <mailto:[email protected]>>> > There are flags to use the data from the calculation which hasn't finished, > check the manual. One of them is UseSaveData, but you can control which > files you wish Siesta to use in continuation to an unfinished job. > > Cheers, > > Marcos > > > On Fri, Jul 23, 2010 at 12:49 AM, Bin Shao <[email protected]<mailto: > [email protected]><mailto:[email protected]<mailto: > [email protected]>>> wrote: > Dear all users, > > I'm a new user of siesta 2.0.2. I'd like to ask how to stop and continue > the calculation, for I submitted a job to a cluster and it hasn't finished > in the request time. > > Thank you in advanced! > > Best, > > -- > Bin Shao > College of Information Technical Science, Nankai University > 94 Weijin Rd. Nankai Dist. Tianjin 300071, China > Email: [email protected]<mailto:[email protected]><mailto: > [email protected]<mailto:[email protected]>> > > > > > -- > Bin Shao > College of Information Technical Science, Nankai University > 94 Weijin Rd. Nankai Dist. Tianjin 300071, China > Email: [email protected]<mailto:[email protected]><mailto: > [email protected]<mailto:[email protected]>> > > > > -- > Bin Shao > College of Information Technical Science, Nankai University > 94 Weijin Rd. Nankai Dist. Tianjin 300071, China > Email: [email protected]<mailto:[email protected]> > > > > -- > Bin Shao > College of Information Technical Science, Nankai University > 94 Weijin Rd. Nankai Dist. Tianjin 300071, China > Email: [email protected]<mailto:[email protected]> > -- Bin Shao College of Information Technical Science, Nankai University 94 Weijin Rd. Nankai Dist. Tianjin 300071, China Email: [email protected]
