Monitoring, recovery, and precise logs can often do more to reduce your number of outages and limit the scope of system failures than the typical panacea more hardware, more software, and more hot spares.
People often equate high availability with redundancy. While redundancy in particular, and technology in general, are necessary components of a highly-available information technology environment, they are hardly sufficient. Here’s a broader perspective on achieving high availability in enterprise-class systems based on the real-world experiences of IBM’s High Availability Center of Competency.