Input: .cif file containing material-of-interest structure, single text file containing what the user would like to calculate and under what conditions.
Output: whatever the user wants, as listed at the end of the input file.
Detailed description: the most basic calculation using the Elk code on our own clusters involves three files: a .in file, which contains information about the calculation that is to be run; a .spacegroup file that contains structural information about the material-of-interest; and a .slurm file to submit the job to the cluster.
Preparing files takes up a lot of time, especially when one wants to run many calculations at once on many different materials. It would be very nice to have a program that only requires two input files. The first of these is a direct .cif file of the structure to be calculated on. The program would read information directly from the cif file and convert it into .spacegroup file. The other file, a .txt file perhaps, would only require the user to put in commands on separate lines, regardless of the order. Furthermore, the program would not require the user to input some of the Elk commands that are usually not changed from calculation to calculation. Furthermore, the calculation would already have all the special points for each Brillouin zone, so all the user has to do is type in what order they would like them (since the .cif file will provide the space group necessary for the program to identify which BZ to use).
Impact: this code would reduce the amount of time spent preparing input files for calculations, as one would simply need to write a simple text with a few lines for the calculation to run, and have the .cif on hand. This would also greatly reduce the learning curve for performing calculations, as well as enable those who are less knowledgeable with DFT to run their calculations easily, without much prior knowledge.
Input:
input.txt
(GitHub does not support upload of .cif files, otherwise would upload Si.cif)
Output:
elk.txt
spacegroup.txt
Input: .cif file containing material-of-interest structure, single text file containing what the user would like to calculate and under what conditions.
Output: whatever the user wants, as listed at the end of the input file.
Detailed description: the most basic calculation using the Elk code on our own clusters involves three files: a .in file, which contains information about the calculation that is to be run; a .spacegroup file that contains structural information about the material-of-interest; and a .slurm file to submit the job to the cluster.
Preparing files takes up a lot of time, especially when one wants to run many calculations at once on many different materials. It would be very nice to have a program that only requires two input files. The first of these is a direct .cif file of the structure to be calculated on. The program would read information directly from the cif file and convert it into .spacegroup file. The other file, a .txt file perhaps, would only require the user to put in commands on separate lines, regardless of the order. Furthermore, the program would not require the user to input some of the Elk commands that are usually not changed from calculation to calculation. Furthermore, the calculation would already have all the special points for each Brillouin zone, so all the user has to do is type in what order they would like them (since the .cif file will provide the space group necessary for the program to identify which BZ to use).
Impact: this code would reduce the amount of time spent preparing input files for calculations, as one would simply need to write a simple text with a few lines for the calculation to run, and have the .cif on hand. This would also greatly reduce the learning curve for performing calculations, as well as enable those who are less knowledgeable with DFT to run their calculations easily, without much prior knowledge.
Input:
input.txt
(GitHub does not support upload of .cif files, otherwise would upload Si.cif)
Output:
elk.txt
spacegroup.txt