Hubble is part of the Cilium platform, and the idea here is to extend that network observability. We like to use the analogy that it's a little bit like Wireshark at the Kubernetes level, so the idea is that it's supposed to be as non-intrusive and transparent as possible to anyone running their workload, but it gives you a higher level of detail than just service discovery, which is one of the great features of Kubernetes. One of the things we can do with Hubble, because we pull the data from the data path of Cilium, is use all that eBPF goodness that we like to talk about to build out really cool functions such as service maps. Once we've got a service map and we understand how your applications operate together, then we can do other things. We can start to think about network monitoring and network alerting as well. With this, we can also do some really cool stuff at the application layer itself. Think about that Layer 7 capability, exposing HTTP metrics. How long did it take for a request to be serviced? Was that a GET against a certain path of your application? Does that take longer than other parts of your application? When we speak to customers and find out what they love about Cilium as a whole, the answer is typically Hubble. They come to us and use Cilium for one of its core functions, maybe service mesh, maybe the BGP features, but then they start to look at the additional parts of that platform and start to consume Hubble. That's the observability capability it has, and it's a thumbs-up from them.