Intel 4th Gen Intel Xeon Scalable processors (source: Intel)
Calling it a “pivotal moment” in its turnaround, Intel today made it official: the company launched its new and much-discussed 4th Gen Intel Xeon Scalable processors (“Sapphire Rapids”), the Xeon CPU Max Series (“Sapphire Rapids HBM”) and the Intel Data Center GPU Max Series (“Ponte Vecchio”).
For HPC, Intel said the new chips “bring a scalable, balanced architecture that integrates CPU and GPU with oneAPI’s open software ecosystem for demanding computing workloads in HPC and AI,” adding that the Xeon CPU Max Series “is the first and only x86-based processor with high bandwidth memory, accelerating many HPC workloads without the need for code changes.”
The Xeon CPU Max Series offers 64 gigabytes of high bandwidth memory (HBM2e) on the package, providing up to 3.7 times more performance than pervious products, according to the company, while the GPU Max Series packs more than 100 billion transistors into a 47-tile package. As a combined platform, the Xeon CPU and GPU Max Series achieves up to 12.8 times greater performance than the prior generation when running the LAMMPS molecular dynamics simulator, the company said.
In AI, compared to previous generation products, 4th Gen Xeon processors achieve up to 10x higher PyTorch real-time inference and training performance with built-in Intel Advanced Matrix Extension (Intel AMX) accelerators, the company said, adding that in natural language processing, Intel said the Xeon CPU Max Series offers up to a 20x speed-up on large language models.
Compared with prior generations, 4th Gen Intel Xeon processors deliver a 2.9x average performance per watt efficiency improvement for targeted workloads when utilizing built-in accelerators, up to 70-watt power savings per CPU in optimized power mode with minimal performance loss, and a 52 to 66 percent lower (TCO), according to Intel.
“The launch of 4th Gen Xeon Scalable processors and the Max Series product family is a pivotal moment in fueling Intel’s turnaround, reigniting our path to leadership in the data center and growing our footprint in new arenas,” said Sandra Rivera, Intel EVP/GM of the Datacenter and AI Group. “Intel’s 4th Gen Xeon and the Max Series product family deliver what customers truly want – leadership performance and reliability within a secure environment for their real-world requirements – driving faster time to value and powering their pace of innovation.”
Intel “Ponte Vecchio” Data Center GPU
On the manufacturing front, Intel said 4th Gen Xeon combines up to four Intel 7-built tiles on a single package, connected using Intel EMIB (embedded multi-die interconnect bridge) packaging technology and delivering increased memory bandwidth with DDR5, increased I/O bandwidth with PCIe5.0 and Compute Express Link (CXL) 1.1 interconnect.
The company said 4th Gen Xeon’s modular architecture means Intel can offer nearly 50 targeted SKUs for customer use cases or applications, including purpose-built SKUs for cloud, database and analytics, networking, storage and single-socket edge use cases. The processors are “on demand-capable and varies in core count, frequency, mix of accelerators, power envelope and memory throughput as is appropriate for target use cases and form factors,” the company said.
“Intel is delivering differentiated solutions and systems at scale to tackle their biggest computing challenges,” the company said. “Intel’s unique approach to providing purpose-built, workload-first acceleration and highly optimized software tuned for specific workloads enables the company to deliver the right performance at the right power for optimal overall total cost of ownership.”
