Tech | Visa | Scholarship/School | Info Place

Behind the Apple lawsuit, chip startup Rivos plans its next move

Rivos made headlines in 2022 after Apple filed a trade secrets lawsuit against Rivos, accusing Rivos of hiring dozens of Apple engineers and using confidential information to develop chips to rival those from the iPhone maker.

Rivers has denied the accusations and countersued Apple for unfair competition. Apple eventually settled the lawsuit in February. Around the same time, it ended separate lawsuits with several Apple engineers employed by Rieves.

Now, with the court drama unfolding, Rivos is redoubling its efforts to bring its chipset technology to market, CEO Puneet Kumar told TechCrunch.

“Rivos was founded with a mission to build industry-leading energy-efficient, high-performance chips,” Kumar said. “We’re excited to target customers who are building data-driven solutions.”

Substantial new funding will help fund these efforts.

Rivos announced on Tuesday that it had raised more than $250 million in an oversubscribed extended Series A round led by Matrix Capital Management and with participation from chip giants such as Intel (through its corporate venture capital arm) and MediaTek. Other backers include Cambium Capital, Hotung Venture Group, Walden Catalyst, Dell Technologies Capital and Koch Disruptive Technologies.

It’s a big shift for Rivos, which was founded in 2021 about a year ago and has been struggling to raise money from investors and hire employees in the shadow of Apple’s lawsuit. In August, Rivos laid off nearly 20 employees, accounting for 6% of its workforce at the time, and was forced to postpone a planned $400 billion Series A round of financing, The Information reported at the time.

Custom server chips

Kumar said Rivos’ long-term goal is to build chips primarily for servers capable of handling intensive data analytics and artificial intelligence workloads, including generative AI workloads.

“Our target customers are building data-driven solutions, such as those that leverage generative AI and data analytics to drive decision-making,” Kumar said. “There are a lot of companies targeting this market; Rivers Rigorous hardware requirements to support AI models and analytics will reshape the enterprise. “

Rivos’ first chipset is built on RISC-V (an open standard instruction set architecture (ISA)).

The ISA is a technical specification underlying each chip that describes how software controls the chip’s hardware. For general-purpose computing, chip design teams typically license existing ISAs from existing vendors such as Arm or Intel. But RISC-V offers an open, royalty-free alternative.

Rivos’ chip features what Kumar describes as a “data parallel accelerator” that accelerates AI and big data-related calculations, essentially a GPU designed for purposes other than graphics processing. It is manufactured using TSMC’s 3nm manufacturing process. In chip manufacturing, “process” refers to the size of the smallest component that can be embedded on a chip.

3nm is considered close to the cutting edge. While Qualcomm, MediaTek, Nvidia and AMD, among others, are expected to use TSMC’s processes in their upcoming chip series, Apple is the only company to use it in its M3 chipset series in 2024.

In addition to building chips, Rivos is also working on developing standalone data center hardware based on the Open Compute Project’s modular standards, which will effectively act as a plug-and-play chip enclosure. Kumar said it is creating a “firmware-to-application” software stack to program the chip.

“Customers’ workloads can be easily deployed on our more efficient hardware but still use their existing models and databases, giving them immediate benefits,” Kumar added.

Rivos, which has yet to generate revenue, plans to make money by charging customers, primarily large data center operators, for its hardware and complementary software solutions. Early investor David Goel said Rivos’ “low-friction” adoption channel is a key differentiator in the highly competitive chip market.

“The Rivos team has masterfully integrated a groundbreaking new RISC-V architecture with creative accelerators to effectively bring this vision to reality,” Goel told TechCrunch. “Their prototype chip provides a strong demonstration of its unique capabilities.”

But is it differentiated enough?

Fierce competition

As the generative AI craze continues, one of Rivos’ potential customers is large tech companies, which are racing to develop their own in-house chips for artificial intelligence and big data analytics.

Google is developing a fifth-generation TPU and recently released Axion, its first dedicated chip for running models. Amazon has several lines of custom chips. Microsoft entered the fray last year with the Azure Maia AI accelerator and Azure Cobalt 100 CPU. Meta is also slowly advancing its own design.

Meanwhile, dozens of startups are vying for a share of the custom data center chip market, which could hit $10 billion this year and double by 2025.

Groq, a company that develops chips to run artificial intelligence models faster than traditional hardware, recently launched a new business unit aimed at enterprise applications and use cases. Tenstorrent, an artificial intelligence hardware startup led by engineering luminary Jim Keller, is looking to build its chipsets into data centers. South Korean fabless artificial intelligence chip company Rebellions has raised hundreds of millions of dollars in funding to increase production of its data center chip Atom.

But it turns out that Nvidia, the current dominant force in the chip field, is difficult to overthrow.

Nvidia briefly became a $2 trillion company this year, fueled by booming demand for GPUs for artificial intelligence training. Wells Fargo Equity Research estimates Nvidia has 98% market share in data center GPUs, with the company’s data center business expected to reach fourth in 2023 as it builds a new unit to design custom chips for cloud computing companies and others. Quarterly growth of more than 400%.

Given the intensity of competition — and the chilling effect Nvidia’s dominance has had on the funding of would-be rivals — times are tough for some of the custom server chip upstarts.

Graphcore has reportedly had $1 billion slashed from its valuation after its deal with Microsoft fell through. A few months ago, the company said it was planning job cuts due to an “extremely challenging” macroeconomic environment. Habana Labs, an artificial intelligence chip company owned by Intel, laid off about 10% of its employees last year. Also last year, SiFive, like RISC-V startup Rivos, laid off 20% of its employees and discontinued its core product line.

So will Rivos perform better? perhaps.

Kumar wouldn’t talk about customers, and Rivos’ chips aren’t expected to be in mass production until sometime next year. But Kumar said Rivos, with 375 employees and hundreds of millions of dollars in the bank, can afford to scale up manufacturing and double down on platform and software engineering.

“The rapid changes in generative AI and its merging with data analytics stacks make it critical that accelerators are easy to program and debug and that data can move seamlessly between CPUs and accelerators,” Kumar said. “Rivos is powered by our ‘recompilation’ rather than reinventing the wheel to meet this need.”

#Apple #lawsuit #chip #startup #Rivos #plans #move

Leave a Reply

Your email address will not be published. Required fields are marked *