Deploying and Securing Multi-Cloud and Edge Generative AI Workloads with F5 Distributed Cloud

Deliver and secure multi-cloud and edge-based generative AI workloads with F5 Distributed Cloud. This lab showcases the deployment of generative AI applications on Google Kubernetes Engine (GKE) and Amazon Elastic Kubernetes Service (EKS). However, enterprises can extend this solution to any major public cloud, edge location, or private data center to align with their infrastructure strategy.

With F5 Distributed Cloud Customer Edge, organizations can seamlessly integrate and secure AI workloads across distributed hybrid environments. F5 Distributed Cloud Regional Edge optimizes traffic management, ensuring secure and high-performance application delivery. Once the application is validated for stability, F5 Data Guard protects sensitive data by redacting personally identifiable information (PII) and other confidential details from AI-generated responses.

This solution is designed for enterprise scalability and flexibility, is capable of supporting key IT ecosystems, including VMware, Kubernetes, Red Hat OpenShift, Nutanix, and other leading platforms. Organizations can deploy AI workloads with confidence, ensuring consistent delivery, security, compliance efforts, and operational efficiency across diverse cloud and edge environments.

Deploying and securing multi-cloud and edge generative AI workloads with F5 Distributed Cloud. This lab demonstrates deployment of Generative AI deploys on GKS and EKS cluster. Using F5 Distributed Cloud Customer Edge configuration, it can able to connect to cluster configuration and application hosted on clusters. 

Using Regional Edge incoming request and response will communicate using F5 Distributed Cloud Regional Edge configuration. Once verifying application running stable. Using Data Guard feature you can be able to hide sensitive information from the response and hide the protect the sensitive information from response of AI workload. 

You'll able to deploy the application on clusters and configure F5 Customer Edge and Regional Edge configuration 

Finally able to test the application verifying the application actually listening to the https load balancer and using data guard feature you'll able to test the response of sensitive information. 

This lab demonstrates the deployment of a Generative AI workload across both GKE and EKS clusters. The front end of the application is hosted on the GKE cluster, while a private LLM is deployed on the EKS cluster.

F5 Distributed Cloud, including Customer Edge and Regional Edge configurations, facilitates connectivity between the application components running on GKE and EKS. Communication between the two clusters is managed through the F5 Regional Edge, ensuring seamless integration.

Once the application is accessible via the F5 HTTPS Load Balancer, the Data Guard feature prevents the exposure of sensitive information in responses. This helps mitigate risks associated with Generative AI unintentionally revealing private data.

This lab highlights the capabilities of F5 Distributed Cloud products in securing and optimizing AI workloads.

In this lab, you'll gain hands-on experience in:

• Setting up an F5 Account and Console Access.
• Deploying a Sample GenAI Application on AWS EKS and GCP GKE.
• Deploying F5 Distributed Cloud Customer Edge on EKS and GKE.
• Verifying the application using HTTPS Load Balancer.
• Securing sensitive data using the Data Guard Feature.

This lab provides the resources from WWT accounts. These resources are ephemeral, once lab completed all the work will get taken out.  This lab primarily requires: 

• AWS Elastic Kubernetes Service (EKS)
• GCP Google Kubernetes Service (GKS) 
• F5 Distributed account. 
• F5 Distributed account namespace. 
• F5 Customer Edge and Regional Edge Services. 
• F5 HTTPS Load Balancer. 
• F5 HTTPS Load Balancer - Data Guard Feature.