Perhaps the hardest single thing to understand about application performance management (APM) is how far we still are from making the most of it.
First, “… this is no longer tactical stuff deep in IT. APM is now seen as a system that needs to be layered in, not just in production, but also in Test and Dev.” John K. Waters attributes those words to John Van Siclen, general manager of Compuware’s APM (application performance management) unit. This strategic opportunity is a considerable challenge for most information technology (IT) departments: we specialize in analysis of technical artifacts and designs. APM might be an organizing principle and touchstone for entire organizations, though, from the top down. Are we ready for that kind of success?
We frequently talk of APM as a sort of negative–it’s not about siloed measurement, it’s not isolated, it’s not yet well-covered by analysts, standardization remains weak–and lose track of the potential APM has to structure and inform operations all the way to the executive level. It’s probably time to declare victory in regard to the technical details that usually consume our attention, and concentrate for a while on the qualitatively different chore of making it indispensable to all levels of decision-maker within the organization.
This is especially valuable at a time when the cloud is shaking up so much of the traditional foundations of datacenters: not only are co-workers bypassing IT to arrange for cloud services on a self-care basis, but IT itself is increasingly basing its operations on or integrating with with cloud offerings. A major accomplishment of APM’s end-to-end orientation is to be able to monitor and manage performance even when it depends on parts not hosted in the datacenter.
There are several dimensions beside the tactical-strategic one in which APM has plenty of room to grow: deployment of business intelligence (BI) APM modules and “real-time” managerial reliance on APM are two major ones.