As the value of data continues to increase, organisations need to have a solid, well-thought-out disaster recovery (DR) strategy in place. It’s the only way to protect and — should the worst happen — recover your data. Enterprises should always operate with the assumption that data is under constant threat, whether from hardware failure, human error, data breach, ransomware attack, or natural disaster. According to the UN Office on Disaster Risk Reduction (UNDRR), the latter is happening more often. Global disaster events have nearly doubled over the past 20 years, so don’t discount the weather when you’re writing and reviewing your plan!
It’s human nature to hope for the best. But in a business context — and particularly when it comes to protecting data, systems, and applications — the smartest approach is always to plan for the worst. In cybersecurity circles, businesses often hire professional hackers to test the robustness of their defences. While we wouldn’t recommend flooding your server room on purpose to see how it might cope with an extreme weather event, there are some practical steps you can take to ensure your DR plan is robust and actionable.
This should include ensuring that you understand which data and apps are needed to keep the business running and ensure the DR plan is built around those as a priority. It should also cover exhaustive checks that identify even the most unlikely data loss scenarios — and include any third-party infrastructure or service providers too. Insight such as this will be invaluable for recovering data, and getting apps and services back up and running with minimal disruption.
It’s interesting to note that the customers who suffered the most significant data loss when OVH’s data centre caught fire were those operating on dedicated bare metal servers. These customers gained improved performance and availability. They also gained a feeling of greater security as they had sole use of their servers. These three points tick some big boxes in terms of protecting data and keeping it available.
So OVH’s bare metal customers could have been forgiven for thinking they had DR covered. However, OVH’s bare metal service didn’t include cloud backup as standard — as it did for customers using virtual machines. This highlights that organisations can never take anything for granted when it comes to disaster recovery. If these customers had developed a thorough DR plan, they should have identified that potential weak link and been able to take steps that would mitigate the risk — however unlikely — that they could lose data stored in OVH’s data centre .
Disaster Recovery and the Cloud
Organisations are sending more data into the cloud and continuing to adopt multi-cloud strategies. While this simplifies many aspects of IT deployment, cuts costs, and improves flexibility, it does create some new system management challenges. Disasters are always going to happen, even within the most robust and advanced hybrid cloud environments.
As the way we work changes, so does DR. SaaS applications, for example, have enabled businesses to scale up remote working during the pandemic. But the data stored on SaaS platforms needs to align with data backup and recovery SLAs. This is important because SaaS vendors are responsible for availability — not data recovery. So, SaaS solutions need data protection.
Businesses can’t take a more relaxed approach to DR because they migrated to the cloud. In fact, the potential complexity of hybrid cloud adoption means that it’s more vital than ever to have a tried, tested, and regularly reviewed disaster recovery strategy in place. With that in mind, it’s essential that organisations periodically and thoroughly test the DR plan. This will ensure each step plays out the way it’s planned, and everyone involved knows the role they play.
Companies need a technology partner that understands the nature of their data and can anticipate potential disaster scenarios. Moreover, organisations need a technology partner that can deliver and implement a solution that enables the business to leverage omnidirectional data recovery, from, to, in, across, and between cloud regions.
The pandemic has pushed organisations to evolve and transform their data use. If this year teaches us anything about data and DR in particular, it’s that we should act on industry-wide advice to ensure we are agile and flexible enough to meet the unexpected head-on.