Intel has been quietly building its AI chip portfolio and recently added the Jaguar Shores AI accelerator to its roadmap in a move that seeks to compete with the likes of Nvidia and AMD.
The Jaguar Shores AI accelerator, unveiled at the recent SC2024 supercomputing conference, is a key component of Intel’s strategy to remain competitive. While details are scarce, Jaguar Shores will most likely be the successor to Falcon Shores, which is scheduled to launch in 2025.
Jaguar Shores could be targeted at AI inference – an area where Intel hopes to make headway against Nvidia and AMD. It remains unclear whether Jaguar Shores will be a GPU or an ASIC, but Intel’s current product roadmap suggests that it could be a next-generation GPU designed for enterprise AI applications.
While Intel has lagged behind its competitors in AI training hardware, the company is now focusing on AI inference, leveraging its Gaudi chips and upcoming technologies like Falcon Shores and Jaguar Shores.
However, to regain its competitive advantage, Intel will want to overcome several technical and organizational challenges.
Intel has faced many setbacks in the AI hardware space, especially in the GPU segment. The company’s previous attempt to expand a GPU, the Rialto Bridge, was cancelled due to a lack of visitor interest, leaving establishments such as the Barcelona Supercomputing Centre in a tricky position.
Intel’s plans for the Falcon Shores GPU have been revised several times, first designed as a built-in CPU-GPU product and now redesigned as a standalone GPU.
Intel has largely ceded the AI training market to Nvidia and AMD, with Nvidia dominating thanks to its GPUs and CUDA software stack. Intel CEO Pat Gelsinger highlighted the company’s remote position in the AI race before his retirement, admitting that Intel lately ranks Nvidia, AWS, Google Cloud and AMD.
Intel is betting on its upcoming 18A production node to give it an edge over competitors like TSMC. The 18A process, which incorporates new technologies such as RibbonFET transistors and rear force supply, promises to improve power and performance.
Via HPCwire
Efosa has been writing about generation for over 7 years, first motivated by interest but now driven by a strong fondness for this field. He has a master’s degree and a doctorate in science, which has allowed him to obtain a solid foundation in the analytical field. thought. Efosa has developed a voluntary interest in generation policy, particularly exploring the intersection of privacy, security and policy. His studies focus on how technological advances influence regulatory frameworks and social norms, specifically knowledge coverage and cybersecurity. By joining TechRadar Pro, in addition to privacy policy and generation, he also focuses on security products B2B. Efosa can be contacted at this email: udinmwenefosa@gmail. com
I’m very excited to get rid of my slow USB 3. 2 SSDs with those faster USB four models.
HPE likely would have beaten Supermicro and Dell to win a billion-dollar AI contract, but it’s not for the Colossus supercomputer.
How to Make an Electric Shaver
TechRadar is from Future US Inc, a leading foreign media organization and virtual publisher. Visit our corporate site.