The El Nino effect dramatically amplified the seasonal rainfall for Chennai causing one of the worst flooding in the southeast (or southeastern) Indian city. The disruption was so bad that it actually made material difference to the quarterly financials of IT giants like TCS. On the other half of the globe southern California started seeing the effects and continues to prepare for “potentially destructive” rainfall.
The question of protecting information infrastructure continues to rise given the increased digitalization of the economy. The good news is natural disasters need not be IT disasters. Our customers are asking us more and more for help to protect their information in a way that gives absolute confidence in the ability to recover. They are also reporting incidents where their investments save the day, not just when a natural disaster hits but also for avoiding IT disasters that can be caused by facility issues, power failures, software bugs or manual errors. The most recent recorded incidents include a major provincial hospital in North China reporting a faulty UPS forced array failure which could’ve caused the loss of valuable patient medical records .Another hardware failure event happened at an IT Service Provider that serves banks and financial institutions in Kentucky where 356TB of data were at risk to be lost forever. How did these companies recover their data? They were smart in the way they planned for a natural disaster. These are the steps they took:
- Formed cross functional teams representing application owners, DBAs and storage architects to review data protection and availability requirements for different workloads. Just a note that a lack of coordination often results in accidental architectures that are hard to maintain and can lead to poor visibility and gaps in protection coverage
- Identified potential disasters, failure modes and response strategies for workloads of varying degree of criticality
- If they were in a natural disaster prone area, they thought about the minimum distance a second datacenter needed to be located for business continuity
- Clarified metrics around both recovery time (RTO) and data loss tolerance (RPO) and the impact those metrics had on their business. They needed to think about what the impact would be of their entire sales force not accessing key applications for an hour or half a day? How would that change if was the end of the quarter or end of the year?
- Mapped protection strategies to the different workloads in a way to justify the ROI from different technologies
- Invested in vendors and technologies that were capable of achieving their data protection objectives for today and that could adapt to their changing business needs