Intel’s Newly Launched Xeon 6 P-Core CPUs Bring Game Changer Priority Core Turbo And SST-TF For GPU-Fed AI Systems

May 22, 2025 at 02:28pm EDT
Intel Xeon Granite Rapids-W CPUs To Feature Support On W890 Platform, Up To 128 PCIe Gen5 Lanes, 8-Channel Memory 1

Intel's new Xeon 6 processors have been launched for the NVIDIA DGX B300 platform, enhancing AI performance by bringing innovative features.

Intel's Latest "Granite Rapids" Xeon 6 P-Core CPUs Come with PCT and SST-TF Combo Along with 30% Faster Memory, Making Sure GPUs Never Wait

CPU-side bottlenecks, which some GPU-powered systems have to face, will no longer throttle the performance with Intel's newly launched Xeon 6 Performance Core processors. Intel has today lifted the embargo on three of its latest Xeon 6 P-core processors, such as the Intel Xeon 6776P, a 64-core/128-thread chip that can easily take care of the demands of high-performance GPUs.

Related Story Intel Announces Leadership With Xeon 6 “Granite Rapids CPUs” In Network Infrastructure; Sees Adoption From Mainstream Telecom Operators

Intel's Blackwell-based DGX B300 system is equipped with the Xeon 6776P, and hence, its GPU will now have access to a powerful CPU that will result in a noticeable performance boost, directly helping users with AI workloads.

These new Xeon SKUs demonstrate the unmatched performance of Intel Xeon 6, making it the ideal CPU for next-gen GPU-accelerated AI systems,

We’re thrilled to deepen our collaboration with NVIDIA to deliver one of the industry’s highest-performing AI systems, helping accelerate AI adoption across industries.

- Karin Eibschitz Segal, Corporate VP and Interim GM, DGC, Intel

With the launch of Xeon 6 processors, Intel introduced two features, which it calls "PCT" and "SST-TF" or Priority Core Turbo and Speed Select Technology - Technology Frequency, respectivelyThe PCT increases the frequency of selected high-priority cores on the CPU during critical workloads, ensuring the GPU gets enough instructions for quicker processing. The SST-TF is a feature that tunes the core frequencies according to the loads, ensuring better performance and efficiency.

Intel claims that the Xeon 6 offers 30% faster memory speeds compared to its predecessors. If you are familiar with AI workloads, then you might be aware of how crucial high-speed and high-capacity memory for faster execution is. With support for MRDIMMs and CXL, AI systems based on the Intel Xeon 6 processors can access high-bandwidth memories to speed up workloads.

The Xeon 6 also offers 20% more PCI-E lanes than its predecessors, which opens up more paths for faster and more data transfer to and from the GPU, ensuring users achieve faster I/O performance. That said, with improvements in multiple areas, the Xeon 6 processors seem a powerful solution for enterprises. Through features such as PCT and SST-TF, the Xeon 6 will speed up large AI models training and inference, helping businesses to be more efficient than before.

News Source: Intel

Follow Wccftech on Google to get more of our news coverage in your feeds.