Gregory Machler of CSO reports for Network World that a growing trend in disaster recovery for cloud providers is the use of load-balanced data centers instead of hot-cold data centers. A load-balanced data center replaces the data base suffering from a disaster with a similar data center that operates at a reduced capacity.

A challenge when using load-based data centers for disaster recovery in cloud computing is tracking configurations of the infrastructure of an application. Why? Each application creates multiple elements within their environment such as server names, open IP addresses, DNS mappings, firewall rules, SAN and NAS configurations and database clusters just to name a few. Also, not only does an application create these elements but each element can exist for different environments such as production, test and development.
How to mitigate this challenge? Centralization of infrastructure configuration metadata. Centralized parameters and versioning of applications can decrease problems within the load-balanced data center and help to return the data center back to a stable state when disaster or errors have occurred.