Intel has introduced a new line of cutting-edge enterprise AI systems aimed at delivering top-notch performance efficiency and cost-effectiveness. The latest innovations, known as Xeon 6 with Performance-cores (P-cores) and the Gaudi 3 AI accelerators, promise enhanced capabilities in handling compute-intensive tasks with remarkable efficiency. The Xeon 6 boosts performance twofold compared to its predecessor, sporting an increased core count, doubled memory bandwidth, and integrated AI acceleration features in each core. On the other hand, the Gaudi 3 is tailored for large-scale generative AI applications, equipped with 64 Tensor processor cores and eight matrix multiplication engines for accelerated deep neural network computations. Additionally, it boasts 128 gigabytes of HBM2e memory for training and inference, along with 24 200 Gigabit Ethernet ports for scalable networking.
Intel has recently partnered with IBM to deploy the Gaudi 3 AI accelerators on IBM Cloud, aiming to drive down total cost of ownership and facilitate the scalability and performance of AI applications. "With our Xeon 6 P-cores and Gaudi 3 AI accelerators, Intel is fostering an open ecosystem that empowers our customers to optimize their workloads with enhanced performance, efficiency, and security," stated Justin Hotard, Intel's Executive Vice President and General Manager of the Data Center and Artificial Intelligence Group.
In addition to these advancements, Intel also launched its next-generation Core Ultra Series 2 chips, codenamed Lunar Lake, earlier this year. These chips introduce key architectural changes such as integrating system memory into the chip and reducing the total number of CPU cores, marking a departure from the traditional x86 approach. With over three times the AI performance of previous generations, the Lunar Lake chips will receive free updates to Microsoft's Copilot+ application, enabling broad application at scale. Michelle Johnston Holthaus, Executive Vice President and General Manager of the Client Computing Group at Intel, highlighted the breakthrough power efficiency and compatibility of x86 architecture, emphasizing the extensive software enablement across CPU, GPU, and NPU. This, paired with the Copilot+ partnership, is set to deliver the most competitive hardware and software offering in Intel’s history.
These new products and improvements are poised to address the mounting competition that Intel faces from rivals like Nvidia, AMD, and Qualcomm. Stay tuned for more updates on Intel's advancements in the tech landscape.
Ranking
