Intel and IBM Launch AI Accelerators to Boost Scalability

Intel and IBM have announced a global collaboration to deploy Intel Gaudi 3 artificial intelligence accelerators on IBM Cloud. The offering, which is set to launch in early 2025, aims to optimize the scalability and efficiency of enterprise AI solutions by providing advanced processing capabilities and support for cloud workloads. 

Gaudi 3 will support IBM's Watsonx AI and data platform, giving customers access to additional resources to scale their AI workloads in hybrid cloud environments. This integration will enable more efficient management of applications and workloads while providing greater visibility and control over the software stack.

The collaboration also includes support for Gaudi 3 within IBM Cloud Virtual Servers for VPC, which will enable x86-based enterprises to run applications faster and more securely. The proposed solution focuses on improving the availability, performance and energy efficiency of generative AI workloads, thereby optimizing the price/performance of model inference.

 “Unlocking the full potential of AI requires an open and collaborative ecosystem that provides customers with affordable options and solutions. By integrating Gaudi 3 AI accelerators and Xeon CPUs with IBM Cloud, we are creating new AI capabilities and meeting the demand for affordable, secure and innovative AI computing solutions,” said Justin Hotard, Intel executive vice president and general manager of the AI and data center group.

The integration of Gaudi 3 into IBM Cloud, according to an Intel press release, will also enable enterprises to adjust computing resources as needed, generating potential cost savings and increasing operational efficiency through a focus on the solution's scalability and flexibility. In addition, optimized performance and security will improve user experience and application execution efficiency.

Despite the significant advances represented by the Gaudi 3 AI accelerators in IBM Cloud, Intel notes that it is important to consider some challenges associated with this initiative. Implementing Intel technologies may involve the need for additional hardware, software or enabled services, which could affect integration and overall project cost.