Hackers versus hurricanes: It's time to begin planning for real disasters
The most common objection I come across when discussing disaster recovery (DR) is, “Oh, well we don’t get hurricanes or floods or anything like that, so we don’t need DR”. To which I always reply “That’s all well and good, but do you have humans? What about hardware and power? Do you have HVAC in your data centre?”
A disaster can be any number of things and yes, whilst nature can definitely be a cause, in research that Opinion Matters undertook on behalf of iland in 2016, we found that it actually accounts for only 20% of outages. Our research also found that more often than not, disaster also comes in the form of operational failures (53%) and human error (52%). It is actually as a result of the latter two factors that the most outages occur.
When people think of disasters, they tend to think dramatically and on a large scale of incidents capable of taking out an entire data centre for long periods of time. In actual fact, when businesses leverage DR solutions, it tends to be because of much more isolated and smaller issues that have impacted the business. Perhaps only a couple of systems or mission critical applications have gone down; entire IT systems don’t need to implode before one’s eyes to warrant a disaster. It is the smaller disasters that we really need to be focusing on as it’s usually a build-up of these which results in a cascading effect which can then have a bigger impact on the business.
We have many customers that will failover just one or two systems. Maybe a patch went wrong, or a virus occurred and the best thing to do was to keep the system running, fail the affected machine over to keep the application running and fail back when systems are repaired. It doesn’t always have to be on a massive scale. By grouping VMs supporting multi-tier applications into virtual protection groups, you can perform partial failovers within that group, reducing cost and simplifying failback.
Significant disruptions to IT systems, however, do happen and can take a heavy toll on the business; hence the importance of having a reliable DR plan in place. Of course, there are the obvious implications of a disaster such as interruptions to trade, service and loss in revenue. But, perhaps what you might not have considered is the effect on employee morale and additionally on customer confidence. In order to reduce disruptions such as these, a proactive and strategic approach to disaster recovery is required.
Five or ten years ago, the solution to DR felt heavy and complex and, as a result, many organisations put it in the ‘too hard or expensive’ basket and DR was put off to next month or the next quarter, maybe even the next year. In the last few years, however, cloud-based disaster recovery options have given organisations new and more affordable route to protecting their businesses.
As companies started to virtualise and people start moving to the cloud, we found that other parts of the datacentre weren’t keeping up. Efficiency was increasing but replication was still lagging, still embedded in the hardware layer. This not only made it difficult to work with an external vendor, as matching hardware requirements could not be met but, it also meant that everything had to be replicated, rather than just the data that companies were interested in. As a result, the investment in digitisation was being undermined as resources were being wasted. iland’s DR software, powered by Zerto, allows for replication to take place in the hypervisor at the VM level, providing its users with a far more efficient solution.
So what should you look for in a DRaaS solution? In my mind, excellent technical support is key in enabling users to craft a complete DR solution that meets their cost requirements and risk tolerance, whilst also ensuring that the solution will be implemented quickly and tailored for success within their organisation. A successful DRaaS solution should also allow for testing by the user themselves to make sure their plan actually works. Finally, a DRaaS solution should be able extend the security measures and compliance rules already present on premise into the cloud.
As I mentioned earlier we commissioned a survey of 250 IT decision makers with responsibility for DR from medium to large companies in the UK. From outage experiences to achievable recovery times to failover confidence levels and barriers to adoption of cloud-based DR, the survey revealed a wealth of insights that can serve as benchmarks for developing a successful disaster recovery strategy. If you would be interested in finding out more about the results of the survey, the full report can be found here.
- » Edge computing and ITOps: Analysing the opportunities and challenges ahead
- » Cloud infrastructure trends: Usage continues to rise – with AWS-VMware workloads soaring in parallel
- » Securing distributed clouds with an integrated approach: A guide
- » Expect 2020 to see public and private cloud outspend traditional IT infrastructure, says IDC
- » Why attaining hybrid IT nirvana means a mix of digital growth and 'digital trust'