How to continue testing with SDV to extend results?

3 visualizzazioni (ultimi 30 giorni)
I am testing a large industrial Simulink model with SDV in order to get the values of structural coverage metics. I originally set the Maximum Analysis Time to be 20 hours. After the 20 hours, the testing stops and I was able to generate a coverage report (progress is shown in first image). However, all of the objectives have not been processed yet, and I want to give the analysis more time to complete. How do I do this without starting the testing over from scratch? Even when I Load Earlier Results that are in the resulting sldvdata.mat file, I am still waiting for the second round of testing to surpass the original objectives processed/satisfied over an hour later (second image). How can I make SDV continue testing where it left off?
Original test progress (20 hours)
Second round of testing
Update: 7+ hours later, my second round of testing is now at the same spot that the original testing stopped at. Obviously it is re-doing some of the analysis to get to this point.
  1 Commento
Pat Canny
Pat Canny il 13 Nov 2020
Hi Monika,
Thanks for your question. I have passed this along to the Simulink Design Verifier engineering team.
- Pat
Simulink Design Verifier Product Manager

Accedi per commentare.

Risposta accettata

Andrew LeCours
Andrew LeCours il 13 Nov 2020
Hi Monika,
If you have SMS, I’d suggest reaching out to MathWorks Support.
1) Thanks for the feedback about the reuse of previous results
A few ways to re-use previous results are the Extend using existing coverage data and Extend using existing test data options. The documentation has a bit more detail on how to use these options.
2) This model seems quite large. Splitting up the model into smaller pieces and generating tests for these smaller chunks can help scale up the Design Verifier analysis.
A few ways to do this are:
a) Run Design Verifier subsystem-by-subsystem.
b) Analyze the model in a bottom-up fashion
You may have already seen this, but this section of our documentation contains more general information about complexity management:
Please let me know if you have any further follow-up questions.
Thanks again for your feedback,
Andrew

Più risposte (0)

Prodotti


Release

R2020b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by