Details
-
Feature
-
Not Assigned
-
None
-
SRCnet
-
-
-
1
-
1
-
0
-
-
-
-
SRCNet0.x example-workflows-and-benchmarks tests-compilation
Description
This feature will adjust each of the tasks in the repo to have 3 definitions of runs - integration test, short run, and long run. The integration test uses a much much smaller data set that completes in 30 seconds, demonstrating the task will run successfully. The short run is expected to complete in <30 mins, and does minimal processing. The long run is anything more substantial, such as a run that takes hours or days. In this way, we can use Makefiles to execute a task with an expectation of how long it will take, e.g. make run-short, make run-long.
This is in preparation for distributing software via containers, where we need more unit and integration tests for all the tasks. Tim Kok has been in contact to look at taking up some of this work.
Attachments
Issue Links
- clones
-
SP-4286 Add additional peak performance benchmark metrics alongside STARS
- Program Backlog