Never miss a story from Verification. Subscribe for in-depth analysis and articles.
The testing and verification of a complex hardware or software system, such as modern integrated circuits found in everything from smartphones to servers, can be a difficult process. One of the most difficult and time-consuming tasks a verification team faces is reaching coverage closure or hitting all events in the coverage space. Coverage-directed-generation (CDG), or the automatic generation of tests that can hit hard-to-hit coverage events and thus provide coverage closure, holds the potential to save verification teams significant simulation resources and time.
Note that the process of just executing more and more runs will continue to (asymptotically) improve coverage and may (perhaps, in a million years) fill all coverage. But coverage maximization is about a smart, automated, much faster way.
With coverage closure being such a stressful time, it’s hard to view it as being more than just a destination. But if we treat coverage closure as a progression instead of just a destination, the path we follow becomes as meaningful as the destination itself.
Using machine learning to efficiently create verification scenarios is a powerful way to enhance advanced verification environments to reduce common verification headaches.
“Do you want better ways to stress important corner cases of your designs? Looking for an extra kick in your regression performance to improve bug exposure rates per $? Harnessing the power of machine learning, which is one of the areas of computational software innovation, Xcelium ML is here to help you optimize your regressions and cover the coverage gaps faster while finding bugs early in your designs.
For example, in a customer’s regression - Xcelium ML learned the regression data and produced a regression with random seeds. This regression ran four times faster than the customer’s regression and hit 99.1% of the target coverage. To hit the last 0.9%, Xcelium ML produced another regression. The total simulation time to achieve 100% coverage by Xcelium ML was 1.7x faster than using the customer’s regression. And, of course, all of this is done while providing the user with comprehensive, diverse analytics to let an engineer in on exactly what’s going on at every step of the process, easing debug.
So, what are you waiting for? Try Xcelium ML today, and don’t waste any more time doing regular regressions, running useless cycles that neither contribute to any increased coverage nor find new bugs. Experience what Cadence’s state-of-the-art computational software can do for your design cycle. For more information, check out the datasheet and learn about how Renesas used Xcelium ML.