"Flow to packet" vs "Find the needle in the haystack"

In an earlier entry, I discussed the importance of using a top-down approach to network and application performance management in order to reduce MTTR. I promised to pick up this theme again and here I am. On one hand, you want to have visibility into the flows instead of looking at all packets if you want to reduce MTTR. On the other hand, you need the capability of diving deeper into the packet data whenever necessary to pinpoint and fix performance problems. In other words, you need a solution that allows you to combine comprehensive real-time and historical monitoring with continuous packet recording and detailed packet level data mining. I will share with you a few customer stories in some key vertical markets in upcoming posts.