F5, Nvidia team to boost AI, cloud security

F5 and Nvidia are expanding their partnership to help enterprises build AI infrastructure and bolster cloud-based application security.

Specifically, the companies will be integrating F5’s BIG-IP Next for Kubernetes platform and Nvidia BlueField-3 DPUs to offer customers a package capable of supporting AI networking and security duties while ensuring traffic management for cloud-based Kubernetes applications. 

F5’s BIG-IP Next application delivery platform features load balancing, application acceleration security, DNS services and application API management capabilities. It integrates with Kubernetes clusters for visibility, control, and Kubernetes application automation. BlueField-3 DPUs deliver higher-speed networking performance compared to traditional CPUs and can offload tasks such as packet processing, routing, and firewalling from core servers.

“The immense data processing requirements of AI place considerable strain on traditional network infrastructure, making it difficult to maintain optimal performance,” wrote Ahmed Guetari, general manager and vice president of products for F5, in a blog about the news.

By offloading and accelerating high-bandwidth network and security tasks—such as packet processing, encryption, and compression—BlueField-3 DPUs deliver cloud network connectivity. This optimization enhances overall performance and accelerates GPU access to data, Guetari wrote.

The idea is to let customers deploy scalable, secure AI solutions faster, with better performance, and while ensuring data remains protected, Guetari wrote.

F5 BIG-IP Next for Kubernetes deployed on Nvidia BlueField-3 DPUs offers a secure and compliant AI networking infrastructure, letting customers adopt advanced AI capabilities without compromising data privacy, wrote Ash Bhalgat, senior director of cloud, telco, and cybersecurity market development for networking at Nvidia,in a blog.

“By offloading tasks like load balancing, routing and security to the BlueField-3 DPU, F5 BIG-IP Next for Kubernetes efficiently routes AI prompts to LLM instances and reduces energy use. This ensures scalable AI performance while optimizing GPU resource utilization,” Bhalgat stated.

In addition, the collaboration aims to help governments and industries manage sensitive data while accelerating AI application delivery. The sovereign cloud market is projected to reach $250 billion by 2027, according to IDC. Meanwhile, ABI Research projects the market for AI foundation models will be $30 billion by 2027.

Sovereign clouds are built to meet strict data privacy and localization requirements. They’re critical for industries handling sensitive data, such as telecommunications and financial services, as well as government agencies, Bhalgat stated.

This isn’t the first time F5 and Nvidia have collaborated. For example, in 2021 they worked to integrate F5’s technology with the Nvidia Morpheus AI framework to offer pretrained AI models and help detect and mitigate security threats.