Open
Description
The WorkflowRequestEvent logic seems dependent on the size of the study sent. This ticket is to review the logic around consuming WorkflowRequestEvent and seeing if there are any potential performance improvements. Details of what was seen during the performance test can be seen below. CT (324 slices) is by far the longest processing time, which conditional logic oly making 1-2 seconds difference, which is expected.
Modality | DICOM Payload Processed | Task Dispatched (WorkflowRequestEvent cosumption) |
---|---|---|
Average (sec) | Max (sec) | Average (sec) |
CT1 ("{{ context.dicom.series.all('0008','0060') }} == 'CT'") | 34.4 | 36.2 |
CT1 ("{{ context.dicom.series.any('0008','0060') }} == 'CT'") | 34.2 | 37.8 |
MR | 1.1 | 1.4 |
US | 1.7 | 2.3 |
RF | 0.7 | 1.2 |
CT (executing Small app or conditional logic) | 34.9 | 37.9 |
RF (no conditional logic) | 0.7 | 0.9 |
Metadata
Metadata
Assignees
Labels
No labels