How to run a Coupled-Cluster calculations in separate steps¶
Typically there are no problems in executing the three steps in a coupled-cluster calculation with RELCCSD: 1) the Hartree-Fock calculation, 2) the MOLTRA step, the integral transformation from atomic to molecular orbitals, and 3) the coupled-cluster calculations itself all in one calculation.
However, in some cases it is not the best approach to perform all these in one go. This is especially true for calculations that use a lot of memory in the coupled-cluster part but are more economical in the previous steps (say, if the system has little or no symmetry or, in Fock-Space calculations, if the active space is large).
In such cases, it is possible to do one step, save the results necessary for the next calculation, and then run the next step. This can considerably shorten the total wall time spent on a parallel calculation, since a larger number of MPI processes can be started for the SCF (and/or MOLTRA) step(s) with the same amount of memory per compute node.
Step 1: SCF¶
From the SCF part, one has to save the DFCOEF file after a successful calculation. If using the 2-component formalism, one can also save the files X2CMAT and AOMOMAT (and possibly AOPROPER), in order to skip the generation of the 4-component to 2-component transformation matrices. If your system supports it, you could keep the scratch directory and then avoid the copying back and forth of these files.
Currently DIRAC’s execution script automatically retrieves these files and places them in a gzip-compressed tar archive, so it is probably not necessary to retrieve the files individually if this archive is kept.
Step 2: MOLTRA¶
MOLTRA will need the DFCOEF file, and also the one-electron Fock matrix (if the 2-component methods are used and also in the case of an frozen density embedding calculation, this will have to be re-generated, so that’s why it is a good idea to keep the X2CMAT and AOMOMAT files around.
The result of the MOLTRA step will be a set of files that start with MDC.... For SCHEME 4 these files can be considered as ‘independent’ of the number of processes used in the parallel run (because only one, created by the master, will have information used by other modules like RELCCSD - such as the number of spinors transformed etc.), so it does not matter how the calculation from which they were obtained was performed.
This is not true for SCHEME 6, because there the information is needed by other modules, and then one has to be careful to use the same number of MPI processes in the MOLTRA step and beyond.
Step 3: RELCCSD¶
RELCCSD will need the MDC* files generated by MOLTRA. One must be careful in the input, however; the keyword .NO4INDEX has to be specified, and the **MOLTRA section must not be present in the input. Again, for SCHEME 6 one must be careful to have the same number of MPI processes as in the MOLTRA step.
A note on Fock-Space Calculations¶
In the case of Fock-Space calculations, an additional trick can be used: if previous/lower sectors are converged, RELCCSD can be restarted and be made skip them. In order to do that one has to do the following apart from the actions outlined above: First, instead of MAXIT=n, use MAXITk=n (where k denotes the sector: 00, 10, 01, 11, 02, 20) and set to zero all the MAXITab for converged sectors to zero. Then restart exactly the same calculation as before (in terms of number of nodes etc.). It is important to have the intermediate files for RELCCSD present in the scratch directory in this case. The easiest way to do so is to keep the scratch directory from the previous run.