Is it possible to make it easier and less risky to integrate IP licensed from an outside group or company?
It’s old news by now that integration of third party is on the rise. And as part of that, engineering teams are constantly trying to figure out how to reduce the risk of using something they’ve licensed from outside their group or company. There may exist a notion that third party IP is more plug and play than it actually is, and there is an increasing burden on the IP provider to ease the integration.
What some designers may not realize what accompanies third party IP is a substantial amount of documentation, software and integration information, which is essential to proper integration of that IP. Still, there is really no way to design an SoC today without IP of some kind.
During last week’s Design Automation Conference, I moderated a panel discussion in the IP Track: Integration Challenges Of 3rd Party IP – Worth The Risk? The panelists hailed from ARM, Cadence, Global Unichip and Synopsys — and each one had very insightful thoughts about this topic, one that I will be exploring in more depth in the coming weeks. From managing data file sizes and IP versions to interoperability testing and the challenge of working with hard IP, these are all topics that engineering teams are dealing with today and deserves in depth analysis.
Let’s start the discussion now. What is your experience of integrating third party IP?