Powering Innovation in the Era of Pervasive Intelligence

In his first SNUG Silicon Valley keynote as CEO and President of Synopsys, Sassine Ghazi outlined the company’s vision to power innovation in our era of pervasive intelligence. Providing a backdrop for the talk is the global semiconductor industry’s tremendous growth. Case in point: while it took the industry 60 years to achieve $500B in sales, it will only take several more years to reach the $1T milestone at the end of the decade.

What is driving this amazing growth?

From smartphones to self-driving cars, the electronic devices and systems that propel our modern world are only getting smarter. Ghazi highlighted three key paradigms in particular that are powering innovation from silicon to systems:

• The rapid acceleration of AI, which is driving dramatic improvements in productivity in many end markets
• The increasing proliferation of silicon, which is powering AI acceleration, bringing it to multiple applications and end markets
• And the increasing importance of software-defined systems, where the interaction of silicon and the system provides the best outcomes for the workloads running on the silicon

Together, these paradigms are shaping the new era of pervasive intelligence, where AI and smart technologies are seamlessly integrated into the fabric of our daily lives, becoming omnipresent and interconnected. While the industry is on a healthy growth trajectory, continued success relies on collaboration between everyone in the ecosystem, from Synopsys to the semiconductor companies and the foundries. 

Changing the World, One System at a Time

This year marks the 35^th anniversary of the Synopsys User Group (SNUG). The excitement was certainly palpable from the standing-room-only audience inside the keynote ballroom at the Santa Clara Convention Center, as Ghazi outlined the incredible trajectory that the industry is on.

“This era is pushing the limits of compute and energy,” Ghazi said. “To address the challenges that it’s bringing, where every aspect of architecting the system to designing the silicon for the software that is tying them together… it requires a new methodology, a new paradigm, a new approach to design that silicon.”

Ghazi traced the different stages of compute, from its introduction in the early ‘70s to the PC era and today’s world of generative AI, where virtually anyone can prompt a model to derive a specific generated outcome from the compute resources. It’s no longer possible to think about silicon before thinking about the overall system. “That’s an unbelievable change that is happening right now. And what is driving it is that accelerated new generation of compute in order to get to that outcome,” he said.

All of this will drive an explosion of intelligent systems, where everyone can essentially be a programmer and everything is programmable. This, in turn, will shape the next 5 to 10 years, helping to drive growth in the industry to reach the $1T mark.

SNUG attendee enthusiasm only grew when Ghazi brought to the stage NVIDIA Founder and CEO Jensen Huang. Ghazi highlighted NVIDIA as an example of a company that has transformed itself to a silicon, systems, and software company.

“Computers…computing technologies…chip technologies are the most important industry in the world and without Synopsys tools, we wouldn’t be able to do what we do at all,” Huang said.

The two leaders talked about how computing has evolved, the new revolution that AI has sparked, and the move from silicon to systems. “Where the chip starts and ends and where the system starts and ends is amorphous,” said Huang. “We need to build the entire chip, which is as far as we’re concerned the entire system, in silicon, in digital twins. When we hit enter, we need to know it’s perfect.”

Huang continued, “We’re all together at the beginning of something very big. This is the first time in 60 years that computing is going to get reinvented, software will be reinvented. Everything about what we build and how we build it will profoundly change in the next several decades. Together we’ll change the world again.”

Empowering Innovation Together

In the second half of his talk, Ghazi highlighted three main challenges the industry must address immediately, in the short term, and in the mid-term: silicon complexity, productivity, and the merging of silicon and systems. While the industry has historically followed Moore’s law to achieve power, performance, and area (PPA) benefits, increasingly complex silicon means that the predictability and ROI are no longer the same in terms of performance/Watt.

However, a broad IP portfolio, an end-to-end, hyperconverged EDA platform, and the most advanced multi-die solutions—all available from Synopsys—can enable the angstrom era and the march to trillions of transistors. IP that’s ready to go when a new standard comes out can help preserve chip design resources, becoming a major productivity enabler as the industry grapples with talent shortages, Ghazi said. Meanwhile, AI-driven design, verification, and data analytics solutions, backed by the assistance of GenAI “copilots” integrated with the tools, can also help close the gaps created by engineering resource limitations. With multi-die designs representing the future in the industry, solutions that support everything from early architecture exploration to die-to-die connectivity can help accelerate heterogeneous integration of dies, he said. Ghazi also announced a new product in the works that will provide AI-driven 3D space optimization, with the promise of up to 10x better productivity.

 “Our mission is to empower you to deliver the unprecedented transformation and deliver your products. It’s not only about Synopsys products,” Ghazi said. “It’s how they come together as part of the ecosystem to empower your innovation and aspiration.”