Announced alongside the company’s MI200-series GPUs in December 2021, the new chips improve upon existing third-generation EPYC SKUs with the introduction of 3D V-Cache, a proprietary die-stacking technology developed in partnership with TSMC.
Courtesy of the addition, Milan-X processors offer three times more L3 cache than previous EYPC chips (768MB) and are said to yield up to 66% greater performance across targeted workloads.
AMD EPYC Milan-X now available
The AMD EPYC Milan-X series is billed as the first ever line of data center processors to utilize 3D die stacking. Essentially, this means CPU components like the logic unit and cache memory are stacked on top of each other, maximizing the use of vertical space rather than expanding the total surface area of the SoC.
The benefit of such an approach is that far more cache memory can be squeezed into the same SP3 socket, leading to significant performance increases across technical computing workloads that rely on large quantities of data, such as computational fluid dynamics and structural analysis.
AMD described these kinds of workloads as essential tools for “companies that must model the complexities of the physical world” to inform the design of innovative new products.
“Building upon our momentum in the data center as well as our history of industry-firsts, 3rd Gen AMD EPYC processors with AMD 3D V-Cache technology showcase our leadership design and packaging technology, enabling us to offer the industry’s first workload-tailored server processors with 3D die stacking technology,” added Dan McNamara, SVP and GM of AMD’s server business.
“Our latest processors provide breakthrough performance for missions-critical technical computing workloads leading to better designed products and faster time to market.”
The new Milan-X processors come in four flavors, with core counts ranging from 16 to 64. The most expensive SKU, the EPYC 7773X, comes in at a price of $8,800 per unit, which is roughly 12% higher than the next most performant model.
However, AMD claims that the performance improvements on offer with Milan-X means companies could run identical jobs on far fewer servers, consuming far less power in the process. Apparently, this could lead to a TCO saving of more than 50% over a three-year period.
Are you a pro? Subscribe to our newsletter
Sign up to the TechRadar Pro newsletter to get all the top news, opinion, features and guidance your business needs to succeed!
Joel Khalili is the News and Features Editor at TechRadar Pro, covering cybersecurity, data privacy, cloud, AI, blockchain, internet infrastructure, 5G, data storage and computing. He's responsible for curating our news content, as well as commissioning and producing features on the technologies that are transforming the way the world does business.