AI from the cloud to the network and to the edge was the persistent theme in an investor webinar hosted by Intel and focused on the company’s Data Center and Artificial Intelligence business unit.
The top order of business was to outline the Xeon roadmap for the next few years, its key product line in a total available market (TAM) that Intel now believes with top $110 billion over the next five years.
Xeon highlights:
Current: 4th Gen Xeon systems are now shipping with all major original equipment manufacturers (OEMs) and original design manufacturers (ODMs), as well as with the top 10 global cloud service providers. Intel claims its 48-core 4th Gen Xeon can deliver a 4X performance gain going head-to-head with a 48-core 4th Gen AMD Epyc CPU on a broad set of deep-learning workloads.
Q4 2023: 5th Gen Intel Xeon Scalable Emerald Rapids is Intel’s next Performance-core (P-core) product. The CPU is already sampling to customers. Volume validation is underway.
First-half 2024: Intel Xeon Processor code-named Sierra Forest, the first Efficient-core (E-core) Xeon processor, will boast 144 cores per socket. The first CPU based on the upcoming Intel 3 process, Sierra Forest hit power-on earlier this quarter with multiple OSes booting in record time (less than a day). It’s on schedule with the first samples already out the door.
A Fast Follow: Intel Xeon Processor code-named Granite Rapids, will arrive hot on the heels of Sierra Forest in 2024. Though a specific launch date has yet to be disclosed, its time-to-market will benefit from sharing a platform with Sierra Forest. Shared IP and technology minimize development and design time.
Some additional highlights:
- Intel invented and is leading the ecosystem in developing a new type of DIMM called Multiplexer Combined Rank (MCR) that lets can achieve speeds of 8,800 mega transfers per second, based on DDR5.
- The Habana Gaudi3 AI accelerator has taped in. Performance details haven’t been shared yet, but Habana Gaudi2 is in the environment and offering 1.8x advantage in throughput-per-watt over a comparable A100 server when running a popular computer vision workload.
- There are 15 new FPGA products scheduled to go through the production release qualification (PRQ) process this year – that’s more new product introductions than ever before in Intel’s FPGA business.
- Intel has made contributions to SYCL, an open C++-based programming model, and acquired Codeplay Software (a leader in the SYCL language and community). SYCL is now included in oneAPI so customers can program and compile across CPUs, GPUs and accelerators from multiple vendors.
“When we talk about compute demand, we often look at the TAM through the lens of CPU units. However, counting sockets does not fully reflect how silicon innovations deliver value to the market. Today, innovations are delivered in several ways, including increased CPU core density, the use of accelerators built into the silicon and the use of discrete accelerators,” stated Sandra Rivera, Intel Executive VP and GM of the Data Center and Artificial Intelligence business unit.