What is the advantage of implementing an active-active firewall cluster that is stretched across separate pods when anycast services are configured?

The ability of a firewall cluster to be deployed in transparent mode is a characteristic of the firewall's operating mode, not a specific advantage provided by stretching the cluster across pods with anycast services.

Anycast services are designed around presenting the same IP/MAC combination from multiple locations, allowing the network to route to the nearest instance, thus configuring different MAC/IP combinations would negate the benefits of anycast.

While an active-active cluster implies load balancing capabilities, the primary advantage of stretching it across pods with anycast is about optimizing traffic locality, not necessarily load balancing between firewalls within a single pod.

With an active-active firewall cluster stretched across separate pods and anycast services configured, each pod's local spine switches are configured to prefer the anycast IP/MAC address that is advertised by the firewall cluster member within their own pod. This "local preference" mechanism ensures that traffic originating within a pod is directed to the local firewall instance, optimizing the traffic path and minimizing inter-pod forwarding, which reduces latency and network congestion.