Release notes
In this second release of Selma we focused on adding a lot of optimization features that improve the execution time when multiple samples are analyzed. Another major change is that the configuration process has been simplified by putting all configuration steps in the same file.
Credit goes to Kjell for guiding us through the development process and to David for diligently testing and providing constructive feedback and suggestions.
Highlights
- The slurm configuration step has been placed in the settings.conf file
- Selma can now run one sample per node with the -s flag, this greatly improves the execution time when more than one sample is run at once
- Selma can also be instructed to use a subset of samples from a master sample file containing two or more samples by using the -e flag
- Custom interval support is now officially supported
- There is now a -m flag that is used to select whether to run in tsd mode or locally