Podcast Summary: "Reshaping Workflows with Dell Pro Precision and NVIDIA RTX PRO GPUs"

Episode: Live from GTC: Securing and Scaling AI Systems with Greg Dalle of F5

Date: March 20, 2026
Host: Logan Lawler
Guest: Greg Dalle, Director of Product Management for AI Solutions, F5

Episode Overview

This episode was recorded live at NVIDIA GTC 2026 and features a discussion with Greg Dalle from F5 about the critical role of application delivery, load balancing, and security in the era of AI-powered data centers (“AI factories”). Host Logan Lawler explores F5’s approach to traffic management and the newly-announced security solutions specifically built for high-performance, GPU-accelerated workflows. The conversation demystifies how advanced infrastructure—powered by Dell Pro Precision workstations and NVIDIA RTX GPUs—transforms both scalability and security for modern AI systems.

Key Discussion Points & Insights

Introduction to F5’s Role in AI Workflows

Timestamps: 00:19–01:11

• Greg Dalle introduces F5 as a leader in bringing traffic management and application delivery solutions to large-scale AI environments.
• F5 plays a foundational role in “AI factories”—data centers tightly integrated with multiple GPUs—by managing how requests are routed and handled across these clusters.

"We are actually showing how to optimize AI factories and AI clouds in general, running traffic management, security, and trying to optimize the number of tokens that these factories can generate."
— Greg Dalle (00:33)

Demystifying AI Factories and Load Balancing

Timestamps: 01:11–02:11

• Logan clarifies “AI factory” for listeners—referring to GPU-powered data centers—and asks Greg for a breakdown of F5’s specific responsibilities.
• Greg explains that F5’s core focus is optimizing AI inference workloads by intelligently routing requests to the right nodes, maximizing efficiency through advanced load balancing.
• Historical context: F5 has three decades of experience in load balancing, now adapting this expertise to unique, AI-driven environments, including LLM (Large Language Model) routing and traffic management based on GPU load.

"We optimize from that point up to the worker nodes where the models run, the application run. And we do load balancing. ... But we optimize that for AI. We do LLM routing, intelligent AI load balancing."
— Greg Dalle (01:33)

Announcements and New F5 Solutions at GTC

Timestamps: 02:11–03:23

• Greg outlines key announcements at NVIDIA GTC 2026:
  • AI-grade Routing & Distributed Inference:
    F5 is showcasing solutions to route user requests efficiently from endpoints all the way to application nodes, including deep LLM routing integrations.
  • DNS Innovations:
    Surprisingly, DNS (Domain Name System) is being revitalized as a cool, foundational technology for AI environments.
  • Security & Guardrails—Including Calypso AI Integration:
    Security is paramount as AI gets widely adopted. F5 is introducing “AI guardrails,” leveraging their recent acquisition of Calypso AI, to demo and implement practical AI security solutions.

"We'll show the integration with LLM routing ... and then very basic stuff like DNS, which is becoming cool again for AI. ... The other thing we are talking about a lot is AI guardrails."
— Greg Dalle (02:29)

Security: Layered Protection for AI Workflows

Timestamps: 03:23–04:42

• Logan probes deeper into F5’s security offerings, noting the trend with AI security at GTC and NVIDIA’s own launches.
• Layered Approach to Security:
  • Traffic Segmentation: Basic but essential—segments traffic between different functions within enterprise clients, like banks.
  • Network Security: Includes firewalling and DDoS protection, designed to operate at the massive scale required by AI workloads.
  • NVIDIA BlueField DPUs: F5 runs their protective software on cutting-edge NVIDIA BlueField 3 (and soon BlueField 4) data processing units, providing hardware-accelerated security at the edge of GPU servers.
  • AI Guardrails & Red Teaming:
    F5 actively inspects inference request/response cycles for threats, responds to attacks, and remediates issues—completing a full-stack approach from network edge to AI application layer.

"We run our software on NVIDIA DPUs, Bluefield 3 for now, Bluefield 4 in the future. ... We have the guardrails and there it's really about looking at what's going on in the inference request responses. ... We mitigate, we remediate attacks."
— Greg Dalle (03:37)

Notable Quotes & Memorable Moments

• On why AI security is vital right now:

  "Security is obviously becoming more and more important as people really adopt AI."
  — Greg Dalle (02:29)

• On F5’s historical transformation:

  "We do load balancing. That's where we started about 30 years ago. But we optimize that for AI."
  — Greg Dalle (01:33)

• On leveraging hardware acceleration for security:

  "If you take stuff like DDoS...we run that at the edge of the servers on the Bluefield DPUs."
  — Greg Dalle (03:37)

Resources & Where to Learn More

Timestamps: 04:42–04:50

• Greg invites listeners to learn more at f5.com/ai.

"The easiest is it's very simple. F5.com/ai."
— Greg Dalle (04:50)

Summary

This episode offers an insider’s look at how F5 and its partners are powering the next generation of AI-enabled data centers. From intelligent load balancing for complex inference workloads to multilayered, hardware-accelerated security, F5 aims to be the connective tissue that keeps large-scale, mission-critical AI deployments both efficient and secure.

Listeners gain practical insight into how Dell Pro Precision workstations with NVIDIA RTX GPUs fit into this picture, supporting the dynamic, always-on world of enterprise AI. If you work in IT, AI, or infrastructure—or just love geeking out about how real-world AI systems stay performant and safe—this is a must-listen episode.