MAVIS has been tested on both unix and linux systems. For the standard pipeline, the validation stage is the most computationally expensive. The memory and cpu requirements will vary with two main factors: the number of structural variants you are validating per job, and the size of the bam file you are validating against.
There are a number of settings that can be adjusted to reduce memory and cpu requirements depending on what the user is trying to analyze. See configuration and settings for more details.
Similar trends were observed for the annotation step (see below) with regards to time elapsed. However the memory requirements remained more constant which is expected since, unlike validation, anntotation does not read more data in for more events.