- Monitoring applications have been performed differently than monitoring the infrastructure
- Application-centric monitoring is where things are moving – how to make the application successful in its environment
- From the businesses perspective the value that IT delivers is embodied in the application portfolio - not the infrastructure. (How often do you say "Wow, the network is fast" versus "Why is PeopleSoft slow?"
- This shift makes the IT operations management group even more important as time goes on
- APM = Application Performance Monitoring
- There is no Business Service Management without a matrix of APM
- We are almost at the point that the application can dictate the underlying infrastructure that it needs to perform appropriately (e.g., re-configure a VM)
- The boundary between application and infrastructure is going away
- You won’t be able to distinguish between virtual and physical applications soon
- You need holistic monitoring of your application stack – there is more than one way to this: 4 perspectives:
- End-user experience monitoring (#1 thing to do – most fundamental task)
- Discovering and Modeling the Application
- Deep dive monitoring (middleware, database, network, off-the-shelf application stacks)
- Ebb and flow of transactions (hard work, embryonic)
The market is moving towards vendors who can provide all of these things (the 4 functionalities + performance management). An integrated, suite approach provides the built-in integration that must be provided via significant integration across “best-of-breed” tools in order to weave together the performance management information together.
This was an interesting session. Especially how there is enough value of putting the 4 items together in a suite coupled with a PMDB to enable a "suite" vendor to succeed. Maybe this is why the identity management "suite" vendors have not succeeded? They haven't figured out the "PMDB" side of things that ties everything together and provides additional added value.
Know what I mean?