In a recent earnings call, AMD revealed plans to bolster its line of EPYC processors with Xilinx’s FPGA-powered AI inference engine.
With the first of these new AI-infused processors slated to arrive next year, the move shows the chip giant is working hard to integrate technologies from its $54 billion acquisition of Xilinx into its existing products. .
Still, recent patents from AMD show that the company is working on several methods to connect AI accelerators to its processors, including using 3D chip stacking technology to do so.
AMD isn’t the first chipmaker to pair its processors with embedded FPGAs, as Intel tried something similar after it acquired Altera in 2015. Despite the company announcing and even demonstrating of a combined CPU + FPGA chip, the end product ended up being more of an experiment that he hasn’t revisited since.
Although AMD hasn’t gone into detail about its future FPGA-infused products yet, connecting Xilinx silicon FPGAs to its processors will likely require some sophistication.
Instead of using standard PCIe lanes and a QPI interconnect to connect FGPA chips to a processor like Intel did, AMD’s patents suggest the company is working on an accelerator port to do this in different ways. .
AMD could use 3D stacking chip technology like it does with its Milan-X processors to deliver performance, power, and memory throughput benefits. However, this approach can present thermal challenges that would reduce performance if the chip is placed too close to compute dies.
Thankfully, we’ll likely learn more about AMD’s upcoming AI-infused processors at the company’s upcoming Financial Analyst Day next month.
Via Tom’s Hardware