At Microsoft Build 2024, Microsoft showcased its ongoing expansion and enhancement of global cloud infrastructure tailored for AI applications. Key highlights include the introduction of Azure virtual machines (VMs) based on the energy-efficient Cobalt 100 processors, delivering up to 40% better performance than previous Arm-based VMs. Microsoft also launched the ND MI300X series VMs, leveraging AMD’s powerful Instinct accelerators for superior cost-performance in inferencing tasks like GPT-4.
The Azure Compute Fleet provisioning service was unveiled to simplify the scaling and management of compute resources across diverse VM types, availability zones, and pricing models. This aims to optimize operational efficiency and flexibility for both AI and general-purpose workloads.
Further advancements include the expansion of Microsoft Copilot in Azure, offering AI-driven assistance for managing cloud and edge operations via natural language queries. This feature is being progressively rolled out to all customers, providing flexibility in access and enhancing productivity through simplified management and security capabilities.
Overall, Microsoft’s commitment to enhancing cloud infrastructure underscores its strategy to empower developers with scalable, performant, and cost-effective solutions for driving AI innovation across industries.
