Report the used random seed for reproducibility

Description

None

Environment

While thinking about how to archive results of student's bachelor theses and master thesis, it occured to me that it would be very helpful for reproducibility if SimuCom (and SimuLizar and EventSim) would report and possible store the random seed used for a particular run. That way, students would just have to specify a model version, Palladio version and seed when archiving results.

Any thoughts on that? How are student results stored in Stuttgart?

Activity

Show:
Steffen Becker
August 13, 2018, 11:51 AM

As this is already possible as part of a much larger functionality, I opt for won't fix.

Steffen Becker
August 13, 2018, 11:50 AM

I think, the whole run should be stored, versioned, etc., including the random seed. This is already supported by standard Eclipse means. In the last tab of the run dialog tabs, configure the run to be persisted in its project instead of using the Eclipse metadata store.

Won't Fix

Assignee

Steffen Becker

Reporter

Anne Koziolek

Labels

None

Requirement Category

None

External issue URL

None

Fix versions

Priority

Minor