Intel Delivers Late Sapphire Rapids Xeons to Argonne Labs • The Register

Intel’s always-lagged Sapphire Rapids Xeon Scalable processors are now available… on the chipmaker’s Dev Cloud. However, anyone looking to actually pick up Intel’s next-gen data center silicon is still out of luck, as it’s becoming increasingly likely that the chips won’t start shipping in bulk until early next year.

During his keynote speech at Intel’s innovation event today, CEO Pat Gelsinger announced that both the Ponte Vecchio data center GPUs and the high-bandwidth memory-enabled Sapphire Rapids CPUs have begun shipping to at least one customer: Argonne National Laboratory. This US government lab hopes to build its Aurora A21 exascale supercomputer using components from Ponte Vecchio and Sapphire Rapids.

“He’s the computer brain behind the Aurora supercomputer, and together we’re shipping these blades to Argonne National Labs right now to build the world’s most powerful supercomputer,” Gelsinger boasted, referring to Sapphire Rapids.

The lab machine was supposed to come online last year and has been repeatedly delayed since 2017 because Intel couldn’t deliver chips on time, including the Sapphire Rapids Xeons. It is not clear how many were actually shipped to Argonne.

For everyone else still waiting for these processors, they might not have to wait much longer to get a taste of the chips, although that won’t happen in their own data centers. Sapphire Rapids is one of several products that Intel is making available in its dev cloud. The service is available in beta for select customers to explore Intel’s chips and software capabilities.

“We’re very excited to have developers stepping on the wheels,” said Lisa Spelman, corporate VP of Intel’s Xeon product line, during a press conference ahead of the innovation. “And this is a unique and rare opportunity to do that before you’ve even come to market or had the opportunity to buy it and bring it into your own data center.”

If you actually want to install these processors in your own racks or aren’t allowed into the dev cloud, you’ll probably have to wait until 2023.

Sapphire Rapids has become a sore point for Intel as the chip’s launch date has steadily climbed slipped backwards. It was supposed to land in 2021, then that was changed to ship to everyone in the first half of this year, then by the end of this year and now probably next year.

If you’re being generous with Intel, consider that Sapphire Rapids is the chipmaker’s most ambitious yet. The Xeons take AMD cues and adopt a multi-die, aka chiplet, packaging technique. Presumably this was done to overcome yield issues and increase core counts, another sore point for Intel, which has lagged its x86 and ARM competitors in core counts for several generations. Sapphire Rapids is also Intel’s first data center chip to support PCIe Gen 5.0, Compute Express Link, DDR5 memory and HBM.

Intel declined to confirm the availability of Sapphire Rapids during the keynote. The registry asked for clarification on the timeframe and was informed that Intel is not releasing launch times just yet. That will fuel industry speculation about a launch in the first quarter of 2023.

In conversation with BloombergSandra Rivera, EVP and GM of Intel’s Datacenter and AI Group, blamed overly ambitious design decisions for the delays.

Sometimes when we find problems it shifts the schedule but we think this is the right compromise for our customers

And in response to questions this month, said Ronak Singhal, senior fellow at Intel The registry Chipzilla prioritizes quality with this generation.

“We will not compromise that quality, even if it means delaying the product,” he said. “We go through a rigorous validation cycle and when we find issues the schedule is sometimes pushed back, but we think this is the right compromise for our customers and what they expect from us.”

Intel has confirmed that at least one SKU will not appear until next year. At this week’s Innovation, Intel shared details of a telecom-optimized Xeon designed to accelerate radio access network (RAN) workloads running at cellular sites.

The chipmaker’s vRAN Boost Xeon Scalable processors include dedicated accelerators for virtual RAN tasks. vRAN enables virtualization of a litany of specialized hardware used to power cell towers, much like appliances like firewalls, routers, and switches have been virtualized in the data center.

By building the accelerators required to virtualize RAN functions into members of its Sapphire Rapids family, Intel can reduce complexity and power consumption by up to 20 percent compared to traditional accelerator cards.

Meanwhile, the data center CPU market is getting crowded. AMD is expected to launch its fourth generation Epyc Genoa This fall, Arm aggressively pushes on his Neoverse V2 Architecture that makes Nivdia’s grace CPU and amps are already working on one follow up to its Altra Max CPUs based on a custom core design.

And when Sapphire Rapids launches in 2023, not only will it face competition from other chipmakers, but also Intel’s upcoming Emerald Rapids processors, which will be available at least from February are on track for a 2023 launch. ® Intel Delivers Late Sapphire Rapids Xeons to Argonne Labs • The Register

Laura Coffey

World Time Todays is an automatic aggregator of the all world’s media. In each content, the hyperlink to the primary source is specified. All trademarks belong to their rightful owners, all materials to their authors. If you are the owner of the content and do not want us to publish your materials, please contact us by email – The content will be deleted within 24 hours.

Related Articles

Back to top button