Inspired by DeepMind's AlphaZero that has mastered complex games like chess and Go, the DSO.ai solution is an AI and reasoning engine capable of searching for optimization targets in very large solution spaces of chip design. The solution, says the company, revolutionizes chip design by massively scaling exploration of options in design workflows while automating less consequential decisions, allowing SoC teams to operate at expert levels and significantly amplifying overall throughput.
"Ever since the introduction of Design Compiler in the late '80s, Synopsys has been enabling silicon innovators with tools and technologies across the design spectrum," says Sassine Ghazi, general manager, Design Group at Synopsys. "With DSO.ai, once again, Synopsys is starting a new chapter in semiconductor design. More than two years ago we set out on a fascinating journey to bring AI to chip design, partnering with academic researchers, industry thought leaders, and AI technology pioneers. Today's announcement marks a very important milestone, and our journey in AI is only just beginning."
A chip design workflow typically consumes and generates terabytes of highly dimensional data compartmentalized and fragmented across many separately optimized silos. To create an optimal design recipe, engineers have to ingest volumes of high-velocity data and make complex decisions on the fly with incomplete analysis, often leading to decision fatigue and over-constraining of their design, says the company.
DSO.ai is designed to revolutionize the process of searching for optimal solutions by enabling autonomous optimization of broad design spaces. DSO.ai engines ingest large data streams generated by chip design tools and use them to explore search spaces, observing how a design evolves over time and adjusting design choices, technology parameters, and workflows to guide the exploration process towards multi-dimensional optimization objectives.
DSO.ai, says the company, uses cutting-edge machine-learning technology invented by Synopsys R&D to execute searches at massive scale, autonomously operating tens-to-thousands of exploration vectors and ingesting gigabytes of high-velocity design analysis data – all in real-time. At