Microsoft launches public preview of Arm-based Cobalt chip: Here’s what you need to know

In a major move to strengthen its cloud infrastructure, Microsoft launched the Azure Maia 100 and Cobalt 100 chips, marking the tech giant’s foray into custom silicon design. The announcement was made during the company’s annual developer conference, Build 2024, where CEO Satya Nadella revealed plans for the public launch of Microsoft’s Arm-based Cobalt CPU.

Enterprises can now access a preview of the new Arm-based Cobalt CPUs through Azure virtual machines, as Nadella confirmed in his keynote. Emphasizing performance improvements tailored to cloud workloads, Nadella emphasized the strategic importance of these chips in Microsoft’s ecosystem.

Also read: MS Dhoni says: “Nothing good has happened on Twitter” when asked why he prefers Instagram

Targeting AI advances

Microsoft’s investment in custom silicon addresses specific needs in its cloud infrastructure, especially around AI capabilities. With its custom AI chip, Microsoft aims to train large language models, potentially reducing reliance on external solutions like Nvidia. Additionally, the introduction of Arm-based CPUs is intended to meet the growing demands of cloud workloads, in line with Microsoft’s forward-looking approach.

Also read: New Delhi-based doctor uses Apple Vision Pro for first bariatric surgery- Know all the details

The Cobalt CPU, which has 128 cores and is built on the CSS Arm Neoverse design, promises to enhance overall cloud services on Azure. Initial testing on platforms such as Microsoft Teams and SQL server shows promising performance results. Microsoft plans to rapidly expand access to virtual machines designed specifically for diverse workloads, ensuring scalability and flexibility for its customers.

Maia 100 AI accelerator

Meanwhile, the Maia 100 AI accelerator, named after a prominent celestial entity, is poised to revolutionize cloud-based AI workloads. Designed for tasks such as large language model training and inference, Maia is expected to support some of Azure’s most demanding AI efforts. Notably, Microsoft’s collaboration with OpenAI has played a key role in perfecting Maia’s capabilities, marking a significant step forward in AI infrastructure development.

Also read: Foldable Honor Magic series may launch in India soon- Know what HTech CEO Madhav Sheth is planning

Manufactured using the TSMC 5nm process and featuring 105 billion transistors, Maia represents Microsoft’s commitment to cutting-edge technology. Although its transistor count is lower than that of competitors like AMD’s Nvidia MI300X AI GPU, Maia’s optimization for Azure’s ecosystem emphasizes Microsoft’s focus on efficiency and effectiveness. productivity in the field of cloud computing.