The Rapid Success Of IP Reuse

The rise of IP use and reuse quickly became a challenge in the early 1990s as designs became more complex. Design teams quickly found themselves struggling to manage not only the vast numbers of IP cores that they were reusing but also versioning, permissions, legal approvals and so on.

In the 1990s, software developers were established users of software configuration management (SCM) tools such as open source RCS/CVS or commercial systems. In fact, SCM systems have become such an integral part of a software development environment that practically no significant software project is even started without a SCM methodology in place.

Hardware designers, however, managed design data in ad hoc, home-grown ways. Time-pressed teams resorted to using spreadsheets to manage much of this, but the complexity of the situation quickly overcame that medium. In an attempt to solve the issue, hardware design teams tried to adapt SCM tools to their flows with mixed results. While it worked for RTL teams, it became a challenge for analog, RF and physical design teams used to dealing with binaries and large file sizes. Hardware design teams were growing and spread across multiple sites, which meant managing complex flows and sharing a large volume of constantly changing data across multiple sites.

This gap in design-management tools between software and hardware engineers inspired the creation of today’s IP design-management (DM) solutions.

The first to catch the wave of providing SoC design and IP design management solutions was ClioSoft, founded by Srinath Anantharaman, who formerly held engineering and management roles at Silvar-Lisco (now Silicon Valley Research), Vantage, and Synopsys. Anantharaman realized that hardware engineers could benefit from a similar design-management approach that had benefitted software engineers. Hardware design flows had some unique requirements not met by SCM tools and hardware designers were not as comfortable with Unix command line tools. Anantharaman wanted to fill this gap by creating a hardware configuration management (HCM) system that would help streamline hardware design teams just as SCM tools had done for software teams. ClioSoft launched in 1997 with their flagship SOS (Save Our Software) design collaboration platform. This was the first product to help manage front-end RTL flows. This helped design teams manage existing IP, which hadn’t been efficiently possible previously. To enable designers, achieve more productivity, ClioSoft partnered with major EDA providers to provide integrations with EDA tools and developed a niche managing analog, RF, mixed-signal design data in addition to digital designs.

After ClioSoft, a few other IP design management companies were formed such as IC Manage, and Methodics. However unlike ClioSoft which built its products from the ground, IC Manage and Methodics built their solutions on top of existing software solutions such as Perforce and Subversion. Today the industry continues to optimize its solutions as design needs change with the ever-changing complexity of chip design.

Approaches and platforms that worked for software and some front-end does not always work for all hardware disciplines in the semiconductor industry. This is because software consists mainly of small text files promoting concurrent development with frequent merges. Hardware developers, on the other hand, deal with large binary files that are not amenable to concurrent changes and merging. Data-management systems built as a layer glued on top of a supply chain management (SCM) system often have to work around the SCM system to speed workspace creation or optimize disk space.

With emphasis on IPs reuse gaining a foothold as companies aspire to leverage existing designs in their quest to be the first in rolling out their SoCs, data management tools also started to look at IP reference and reuse flows, making it easier for designers to reuse their IP. Companies achieved ISO 26262 certification and bought an element of traceability as well making it easier for designers to see who is using the IP. Moreover, they also integrated their products with a number of issue tracking tools to enable design teams to be able to review the issues associated with an IP and when fixed, be notified if the team was using the IP.

These kinds of optimizations will continue for the foreseeable future, especially the number of cores in each SoC soars well past the 120 mark. Without IP reuse and effective design-management tools, SoC scaling simply doesn’t work.

But with the growing popularity of IPs also came the problems of IP theft. IP providers began to be wary of their designs being copied by others or being used in violation of the licensing agreements. Large IP companies started to enforce audits and found a number of companies in violation. To circumvent this problem some semiconductor companies started to build their own IP management systems. The only problem with solutions such as these was that the solution was not scalable.

Companies such as ClioSoft provided a paradigm shift by redefining IP to not only include semiconductor IPs, but also flows, documents, scripts etc. and used crowdsourcing as a vehicle to build an IP reuse ecosystem for an enterprise. The designHUB platform continues to help companies leverage their internally developed IPs in a better way and enabled designers to browse, search, compare and reuse IPs with ease and then publish either the entire design or a part of it. It also provided a bridge between the IP developers and IP users by providing a communication channel which eventually became part of a live and growing knowledgebase which others could leverage upon. To resolve issues such as third party IP liabilities, they introduced the notion of workflows which required appropriate approvals prior to download.

IP reuse is here to stay. With different industries such as automotive, IoT, defense, bio medical having their own criteria for IPs, IP reuse platforms will continue to evolve to scale up to meet the needs.