The role of performance monitoring within financial services is multifold. We have the capacity planning arm which is critical. Being able to visualize why something has increased in use and then being able to tell the underlying cause of that could, for example, allow you to retune an application so you didn't have to perform a costly bandwidth upgrade because you can visualize performance and flow data within one window. From an operational perspective, having that 10,000 foot map that we discussed, giving your operators the best clue on reducing the size of the haystack is absolutely critical.
Then above, over and beyond that, starting to move this out to more of a service based view. End to end service based monitoring has always been tricky in large, heterogeneous, multi-zillion path environments, how do you map a path through the network, it's not straightforward. That's another that SDX is going to really complicate because you just started a whole another layer of complexity only that and that's going to be a challenge.