The growing demand for AI-powered solutions has led to the emergence of specialized cloud infrastructures optimized for AI workloads. This talk introduces the concept of "AI Factories"—dynamic, open source-driven architectures designed to streamline the processing of complex AI workloads in a scalable, multi-tenant environment.

We will explore how open source technologies like OpenNebula enable the seamless integration of advanced features, including GPU passthrough and SR-IOV for high-performance compute capabilities, hybrid cloud deployment for scalability and flexibility, and pre-configured AI appliances to accelerate development pipelines. Together, these components deliver an “AI as a Service” model, providing users with a robust and vendor-neutral foundation for training, deploying, and managing AI workloads efficiently.

Through practical examples and architectural insights, attendees will gain a clear understanding of how to build and deploy these AI Factories using open source tools. The session will also highlight the role of hybrid cloud strategies in balancing on-premises and public cloud resources, showcasing how to create cost-effective, high-performance environments tailored for AI innovation.

Join us to discover how to harness open source technologies to power the next generation of AI services.