A brief introduction; I’m Dave Leith, I work in the Technical Solutions dept. at uptime software and have been involved heavily in the Client Services and Sales sides of uptime for the past few years.
As Alex has spoken to on a few occasions in his blog, up.time MDC (multi-datacenter) functionality has helped a number of our users achieve enterprise wide visibility into their global IT Services monitoring and performance health by combining many distinct up.time installations, called Local Datacenters (LDC), into one global Enterprise Management Server (EMS). This mapping of one EMS to many LDCs is the typical approach to utilizing our loosely-coupled MDC architecture.
This week I worked with a client who has turned the tables on our MDC functionality. Instead of providing a single EMS to their users they decided to setup several EMSs all hooked into a collection of LDCs, a many to many relationship as seen below. Why did they do this? In this client’s case, as with many global organizations, each of their datacenters globally serve a set of applications to their globally dispersed lines of business (LOB). By providing one EMS to each line of business the administrators for that LOB can build a dedicated dashboard for their end users showing the global availability of their own IT Services without having to worry about impacting the other LOBs. Each LOB is completely contained from the others so that they can work freely while still having access to all of the raw data from the LDCs that the other LOBs do. From a capacity planners point of view, this allows capacity reporting against a sandbox of historical performance data only includes their own servers and applications performance data, making trending and forward planning much more straight forward.
There are a number of other interesting applications for this many to many MDC relationship which I’ll talk about in future posts.
ShareThis
Subscribe to this post's comments
