Advanced Computing in the Age of AI | Wednesday, February 28, 2024

Nvidia-Arm Deal a Boon for RISC-V? 

If and when Nvidia closes its $40 billion acquisition of Arm Ltd., the transaction could provide a boost for the competing RISC-V architecture.

As regulators in the U.S., China and the European Union begin scrutinizing the impact of the blockbuster deal on semiconductor industry competition and innovation, the deal has at the very least shown a light on the open-source RISC-V architecture and its growing ecosystem. To that extent, it could become a viable alternative for a growing number of hardware acceleration use cases.

As we have reported, market analysts note that Nvidia also has a stake in the RISC-V architecture, and leveraging its ownership of Arm against a rival chip architecture would ultimately prove self-defeating.

The organization behind RISC-V has attempted to play down the Nvidia-Arm deal, which was announced Sept. 14, noting the chip sector is accustomed to disruption, which can fuel competition that brings chip innovations.

“With any disruption, there are discussions about the ripple effects on the industry,” Calista Redmond, CEO of RISC-V International, noted in a blog posted shortly after the Nvidia-Arm deal was announced.

“Many people are wondering whether companies will be investing more heavily in RISC-V as a result of our open model for licensing and collaboration or due to the potential limits placed by companies or nations on proprietary architectures,” Redmond added.

The RISC-V ecosystem has grown to about 200 companies from 50 countries, and Redmond asserted investment would continue to grow whether or not the Nvidia-Arm deal is approved by regulators.

Given the recent surge in semiconductor industry growth, up more than 23 percent from the same period last year, the chip sector likely has room for competing Arm-based and RISC-V chip architectures. Among the reasons are growing demand for AI accelerators and the rollout of 5G services. The next-generation wireless technology with its larger, low-latency data pipeline would yield direct performance benefits for cloud and edge computing, data analytics, the internet of things (IoT) and machine learning, observers note.

“Many companies will look more closely at RISC-V for next generation designs, if they aren’t already,” Redmond argued.

Indeed, RISC-V was front and center during this week’s Linley Fall Processor Conference. In a presentation on application-specific hardware accelerators, chip analyst Gwennap Linley flagged several approaches for boosting microcontrollers dubbed “tiny AI engines,” so-called because they can be embedded in low-power IoT devices.

Among those mentioned was a nine-core accelerator from GreenWaves Technologies based on the RISC-V architecture. Power consumption for the GreenWaves microcontroller that implements the RISC-V design runs as low as 5 mW, Linley noted.

Other chip designers at the Linley forum including Think Silicon detailed RISC-V GPU designs for AI, computing and graphics workloads that would compete directly with Nvidia and its Arm subsidiary.

About the author: George Leopold

George Leopold has written about science and technology for more than 30 years, focusing on electronics and aerospace technology. He previously served as executive editor of Electronic Engineering Times. Leopold is the author of "Calculated Risk: The Supersonic Life and Times of Gus Grissom" (Purdue University Press, 2016).