TAIPEI, Taiwan, June 4, 2024 — Today at COMPUTEX, Intel has unveiled cutting-edge technologies and architectures poised to dramatically accelerate the AI ecosystem – from the data center, cloud and network to the edge and PC. With more processing power, industry-leading energy efficiency and low total cost of ownership (TCO), customers can now seize every opportunity in the AI system.
“AI is driving one of the most significant eras of innovation the industry has ever seen,” said Pat Gelsinger, CEO of Intel. “The magic of silicon is once again enabling exponential advances in computing that will push the limits of human potential and propel the global economy for years to come. »
Gelsinger continued: “Intel is one of the only companies in the world innovating across the entire spectrum of AI market opportunities – from semiconductor manufacturing to PC systems, networking, edge and data centers. Our latest Xeon, Gaudi and Core Ultra platforms, combined with the power of our hardware and software ecosystem, deliver the flexible, secure, durable and cost-effective solutions our customers need to maximize the immense opportunities ahead.
Intel enables AI everywhere
During his COMPUTEX keynote, Gelsinger highlighted the benefits of open standards and Intel’s powerful ecosystem that help accelerate AI opportunities. He was joined by industry luminaries and leading companies who voiced their support, including Jason Chen, Chairman and CEO of Acer, Jonney Shih, Chairman and CEO of ASUS, Satya Nadella, Chairman and CEO of Microsoft and Jack Tsai, president of Inventec, among others.
Gelsinger and others have made it clear that Intel is advancing AI innovation and delivering next-generation technologies sooner than expected. In just six months, the company went from launching 5th generation Intel Xeon processors to introducing the first member of the Xeon 6 family; from previewing Gaudi’s AI accelerators to offering enterprise customers a cost-effective and efficient Generative AI (GenAI) training and inference system; and from the advent of the AI PC era with Intel Core Ultra processors in more than 8 million devices to the unveiling of the next client architecture scheduled for release later this year.
With these developments, Intel is accelerating execution while pushing the boundaries of innovation and production speed to democratize AI and catalyze industries.
Data Center Modernization for AI: Intel Xeon 6 processors improve performance and power efficiency for high-density, scalable workloads
As digital transformations accelerate, businesses face increasing pressure to refresh aging data center systems to achieve cost savings, meet sustainability goals, optimize space physical on the floor and in the racks and create entirely new digital capabilities across the enterprise.
The entire Xeon 6 processor platform and family is specifically designed to address these challenges with E-core (Efficient-core) and P-core (Performance-core) SKUs to meet the wide range of cases utilization and workloads, AI and other high-performance computing needs for scalable cloud-native applications. E-cores and P-cores are based on an architecture compatible with a shared software stack and an open ecosystem of hardware and software providers.
The first of the Xeon 6 processors to debut is the Intel Xeon 6 E-core (codenamed Sierra Forest), available starting today. The Xeon 6 P-cores (codenamed Granite Rapids) are expected to launch next quarter.
With high core density and exceptional performance per watt, the Intel Xeon 6 E-core delivers efficient computing with significantly lower power costs. Improved performance with increased power efficiency is perfect for the most demanding high-density and scalable workloads, including cloud-native applications and content delivery networks, network microservices, and consumer digital services.
Additionally, the Up to 2.6x performance gain per watt compared to 2nd Gen Intel Xeon processors on media transcoding workloads1. Using less power and rack space, Xeon 6 processors free up compute capacity and infrastructure for innovative new AI projects.
Delivering high-performance GenAI at significantly lower total cost with Intel Gaudi AI accelerators
Today, harnessing the power of generative AI is becoming faster and less expensive. As the dominant infrastructure choice, x86 operates at scale in almost any data center environment, serving as the foundation for integrating the power of AI while ensuring cost-effective interoperability and the enormous benefits of ‘an open ecosystem of developers and customers.
Intel Together, these two elements provide a powerful solution that integrates seamlessly with existing infrastructure.
As the only MLPerf-referenced alternative to the Nvidia H100 for large language model (LLM) training and inference, the Gaudi architecture provides customers with the GenAI performance they seek with a price-performance advantage that provides choice and rapid deployment time at a lower total cost of operation.
A standard AI kit comprising eight Intel Gaudi 2 accelerators with a Universal Base Board (UBB) offered to system vendors at $65,000 is estimated to be one-third the cost of comparable competing platforms. A kit including eight Intel Gaudi 3 accelerators with UBB will cost $125,000, two-thirds the cost of comparable competing platforms4.
Intel Gaudi 3 accelerators will deliver significant performance improvements for training and inference tasks on leading GenAI models, helping businesses unlock the value of their proprietary data. Intel Gaudi 3 in a cluster of 8,192 accelerators is expected to deliver up to 40% faster training time5 compared to the equivalent sized Nvidia H100 GPU cluster and up to 15% faster training throughput6 for a cluster of 64 accelerators compared to the Nvidia H100 on the Llama2-70B Model. Additionally, Intel Gaudi 3 is expected to deliver up to 2x faster inference on average7 than Nvidia H100, running popular LLMs such as Llama-70B and Mistral-7B.
To make these AI systems widely available, Intel is collaborating with at least 10 of the world’s leading systems vendors, including six new vendors that have announced Intel Gaudi 3. Today’s new collaborators include Asus, Foxconn , Gigabyte, Inventec, Quanta and Wistron, expanding production offerings from major system suppliers Dell, Hewlett Packard Enterprise, Lenovo and Supermicro.
About Intel
Intel (Nasdaq: INTC) is an industry leader, creating revolutionary technology that enables global progress and enriches lives. Inspired by Moore’s Law, we continually work to advance semiconductor design and manufacturing to solve our customers’ greatest challenges. By integrating intelligence across the cloud, network, edge, and every computing device, we unlock the potential of data to transform businesses and society for the better.
Source: Intel
