F5 and NVIDIA have collaborated to enhance the performance, multi-tenancy, and security of AI applications by integrating F5 BIG-IP Next for Kubernetes with NVIDIA BlueField-3 Data Processing Units (DPUs) and the NVIDIA DOCA software framework. This collaborative effort is highlighted by Sesterce, a leading European operator specializing in next-generation infrastructures and sovereign AI, which has successfully validated the solution across several key capabilities. One of the primary benefits of this integration is a significant improvement in GPU utilization, initially showing a 20% enhancement. The solution leverages NVIDIA Dynamo and the KV Cache Manager to reduce latency and optimize GPU and memory resources, which is crucial for the reasoning of large language model (LLM) inference systems. Smart LLM routing on BlueField DPUs allows for efficient distribution of tasks to different models based on their computational requirements, ensuring that simpler tasks are handled by less expensive, lightweight models while complex queries are directed to advanced models. This intelligent routing not only improves output quality but also enhances the overall customer experience by lowering latency and improving the time to the first token. The integration also includes robust scaling and securing of the Model Context Protocol (MCP), an open protocol developed by Anthropic. By acting as a reverse proxy, F5 technology enhances security capabilities for MCP solutions and the LLMs they support. Additionally, the powerful data programmability enabled by F5 iRules allows for rapid customization and adaptation to evolving AI protocol requirements, providing enhanced protection against emerging cybersecurity risks. Youssef El Manssouri, CEO and Co-Founder at Sesterce, noted that the integration of F5 and NVIDIA was promising even before testing. Sesterce’s results have confirmed the benefits of F5’s dynamic load balancing in managing high-volume Kubernetes ingress and egress traffic in AI environments. This approach optimizes GPU usage, bringing additional value to Sesterce’s customers. El Manssouri is optimistic about future innovations from F5 and NVIDIA, particularly in supporting next-generation AI infrastructure. Kunal Anand, Chief Innovation Officer at F5, emphasized that routing and classifying LLM traffic can be computationally intensive, often degrading performance and user experience. By programming routing logic directly on NVIDIA BlueField-3 DPUs, F5 BIG-IP Next for Kubernetes provides an efficient solution for delivering and securing LLM traffic. Anand believes this is just the beginning and looks forward to deepening co-innovation with NVIDIA as enterprise AI continues to grow. NVIDIA, represented by Ash Bhalgat, Senior Director of AI Networking and Security Solutions, highlighted that the combined F5 and NVIDIA solution offers a single point of control for efficient traffic routing to AI factories. This not only optimizes GPU efficiency but also accelerates data ingestion, model training, inference, and other AI processes. Bhalgat praised F5’s support for multi-tenancy and enhanced programmability with iRules, which make it well-suited for continued integration and feature additions. Greg Schoeny, SVP of Global Service Provider at World Wide Technology, commented on the increasing reliance on MCP deployments for agentic AI. He noted that the advanced traffic management and security features provided by F5 and NVIDIA in Kubernetes environments are setting new standards in the industry, offering integrated AI feature sets and automation capabilities that are unmatched. Sesterce, founded in 2018, is a leading European operator focused on high-performance computing and AI infrastructure. The company delivers flexible, sovereign, and sustainable solutions tailored to startups, large enterprises, and academic institutions, aiming to become the European leader in AI infrastructure. Sesterce’s AI-native service layer supports data preparation, VLLMs, and modular business intelligence solutions, all while ensuring privacy and compliance with European standards. F5, Inc. (NASDAQ: FFIV) is a global leader in application delivery and security. With over three decades of expertise, F5 has developed the F5 Application Delivery and Security Platform (ADSP) to deliver and secure applications and APIs in various environments. F5 is dedicated to innovation and partnership, helping organizations achieve fast, available, and secure digital experiences. Industry insiders believe that the collaboration between F5 and NVIDIA is a significant step forward in the development of AI infrastructure, addressing the critical challenges of performance, multi-tenancy, and security. This partnership is expected to drive further advancements in AI technology and provide a robust platform for organizations to build and deploy AI applications efficiently and securely.