There is one simple reason why data center operators care about availability: It is the measure of how much time per year their ICT resource is operational and available.
Availability is defined as MTBF (Mean Time Between Failures) / MTBF + MTTR (Mean Time To Repair).
Our colleagues at Uninterruptible Power Supplies Ltd. published an article on their website, that shows how availability can be increased: by reducing MTTR as well as by increasing MTBF.
But how can this be done in practice? MTBF can be increased in two ways: First, with N+n redundant systems. Second, with the best quality of the components used.
And what about reducing MTTR?
As an answer to this question UPSL provides a table that shows the effects of redundancy and modularity on MTTR and availability. The results are stunning and help to understand why a modular approach to UPS is a big advantage when it comes to increasing availability: