An Introduction to Real-Time Discovery in the Cloud
Change is good right? This isn’t something a typical configuration manager might think when looking at the rate of change you are expected to manage with Cloud Computing. Capturing quick-moving changes is difficult, especially if your discovery occurs on a scheduled basis.
The normal use case for a Configuration Management System (CMS) has been to detect and manage servers and software that are stable and experience little change over time. You rely on this stability so that in large environments, it’s feasible to examine all your systems using a scheduled timetable of scans (i.e. once per day or once per week).
With Cloud Computing, you must now assume that the environment is changing all the time, at any time. You can no longer afford to discover on a schedule. If you don’t dynamically monitor the environment for change in real-time, it is likely you will miss a server being started up and then removed a few hours later.
Think of “Black Friday” in the US for instance, which has the biggest sales in stores or online for the entire year. A retail company may need to spin up new computing resources and software very quickly to react to new demand, and then remove those resources as soon as it’s no longer required. In this scenario, the tools you use to “discover” these new servers must react as quickly to changes the cloud as the cloud itself can react.
Read the entire article here, An Introduction to Real-Time Discovery in the Cloud
via the fine folks at HP Enterprise.