SiFive unveils RISC-V chip design for high-performance AI workloads

SiFive, a designer of chips based mostly on the RISC-V computing platform, introduced a collection of recent AI chips for high-performance AI workloads.

The SiFive Intelligence XM Sequence is designed for accelerating excessive efficiency AI workloads. That is the primary mental property from SiFive to incorporate a extremely scalable AI matrix engine, which accelerates time to marketplace for semiconductor corporations constructing system on chip options for edge IoT, client units, subsequent technology electrical and/or autonomous automobiles, information facilities, and past.

As a part of SiFive’s plan to help prospects and the broader RISC-V ecosystem, SiFive additionally introduced its intention to open supply a reference implementation of its SiFive Kernel Library (SKL).

The announcement was made at a SiFive press occasion, Tuesday, in Santa Clara, the place executives mentioned the management function the RISC-V structure is enjoying on the core of AI options throughout a various vary of market leaders, and offered an replace on SiFive’s technique, roadmap and enterprise momentum.

The open resolution

Patrick Little, CEO of SiFive, mentioned in an interview with VentureBeat that prospects within the semiconductor, programs and client markets have come to understand the software program technique behind SiFive and RISC-V.

He famous that merchandise with greater than 10 billion SiFive cores have shipped so far. And Little famous that SiFive has invested greater than $500 million in R&D and it’s promoting to the highest semiconductor leaders and hyperscalers. The corporate has greater than 400 design wins.

The RISC-V structure has a software program that’s an open commonplace interface, which means any sorts of cores that hook up with it. Which means prospects who use SiFive designs can select their very own accelerators for AI and different purposes with out having to fret about breaking software program compatibility, Little mentioned.

Whereas massive leaders in AI like Nvidia can use their very own proprietary graphics processing unit (GPU) architectures, smaller corporations use their very own breed of accelerators, he mentioned. However software program programmers don’t need to study a brand new language each time a brand new accelerator comes alongside, Little mentioned. So the hyperscalars and chip corporations need to use RISC-V options like SiFive in order that they don’t should maintain rewriting their software program, he mentioned.

The RISC-V open commonplace software program interface permits for the swish evolution of the RISC-V commonplace over time and it de-risks the answer past a single proprietary vendor.

SiFive has been steadily transferring up a meals chain, beginning within the Nineteen Nineties with embedded cores and including its first vector processor in 2021. And now it’s including AI options. Clients can use it as a knowledge move processor because the entrance finish to their processor to go together with their altering backend AI accelerators.

“They don’t need to maintain writing to the AI software program. So we put a RISC-V vector processor in entrance of that. The AI processors maintain altering quick. The fashions maintain altering. Software program writers need to write to one thing that might be round in 15 years,” he mentioned. “We’re one of many few corporations that may fill that hole. And as we speak we introduced personal accelerator, or matrix multiplication engine, and we’re doing the XM product line to completement what we did in vector processing. It’s a matrix multiplication engine.”

Clients who need a substitute for Nvidia can flip to a different supply, however they don’t need that rival to be one other proprietary resolution. Slightly, they like RISC-V because it provides many rival corporations behind it, Little mentioned.

“We imagine our resolution can scale to Nvidia degree efficiency,” he mentioned.

“Many corporations are seeing the advantages of an open processor commonplace whereas they race to maintain up with the speedy tempo of change with AI. AI performs to SiFive’s strengths with efficiency per watt and our distinctive capability to assist prospects customise their options,” mentioned Little. “We’re already supplying our RISC-V options to 5 of the Magnificent 7 corporations, and as corporations pivot to a ‘software program first’ design technique we’re engaged on new AI options with all kinds of corporations from automotive to datacenter and the clever edge and IoT.”

SiFive’s new XM Sequence provides an especially scalable and environment friendly AI compute engine. By integrating scalar, vector, and matrix engines, XM Sequence prospects can make the most of very environment friendly reminiscence bandwidth. The XM Sequence additionally continues SiFive’s legacy of providing extraordinarily excessive efficiency per watt for compute-intensive purposes.

“RISC-V was initially developed to effectively help specialised computing engines together with mixed-precision operations,” mentioned Krste Asanovic, SiFive chief architect, in an announcement. “This, coupled with the inclusion of environment friendly vector directions and the help of specialised AI extensions, are the the reason why most of the largest datacenter corporations have already adopted RISC-V AI accelerators.”

As a part of his presentation, Asnovic launched extra particulars on the brand new XM Sequence which broadens its Intelligence Product household. The XM Sequence additionally continues SiFive’s legacy of providing extraordinarily excessive efficiency per watt for compute-intensive purposes.

That includes 4 X-Cores per cluster, a cluster can ship 16 TOPS (INT8) or 8 TFLOPS (BF16) per GHz. The chip has 1TB/s of sustained reminiscence bandwidth per XM Sequence cluster, with the clusters with the ability to entry reminiscence through a excessive bandwidth port or through a CHI port for coherent reminiscence entry. SiFive envisions the creation of programs incorporating no host CPU or ones based mostly on RISC-V, x86 or Arm. The corporate is sampling its options now.

SiFive might be on the RISC-V Summit North America, happening Oct. 22-23, 2024 in Santa Clara, California. The corporate has 500 individuals.

“We’ve turn into the gold commonplace of RISC-V,” Little mentioned.