Most businesses have backup facilities in place to help them in the event of a data breach or physical disaster that renders their offices or data unusable. But how many know that they can retrieve that data and have their business up and running again in minutes?
Server room floods, ransomware, fires – however your data is damaged, lost or digitally encrypted – do you know how quickly you can retrieve it or even if you can? iland found in a recent survey that just 50% of businesses are testing their disaster recovery (DR) plans only annually or at less frequent intervals, while seven percent did not test their DR at all. Of the organizations testing less frequently, half said their disaster recovery plan may be inadequate based on their most recent DR test, while 12% encountered issues that would result in sustained downtime. Zero respondents said that their DR test was completely or moderately successful. Everyone reported experiencing issues.
So, with most companies remaining badly behind the curve, what steps are needed to ensure that you can retrieve your data after a data breach or disaster?
Understanding your data
The datasets of organizations are huge, but the ability to retrieve 100s of terabytes in minutes is like having a spare car in your garage just in case your main one doesn’t work – it’s expensive to have it all waiting on the off chance you need it. And the faster you need it back, the more it costs.
Therefore, a core aspect of a DR strategy is to prioritize the data that is most critical to the business and focus your efforts around protecting that data first. To understand your data, look at your entire estate and define what’s critical to your business operations. Prioritize it in order of how it would impact customer delivery most if lost. It will give you a focus, and in turn, you can develop measures to minimize data loss in the event of a cyber-attack or disaster. You can also catalog it by how much data can be lost by invoking a recovery (RPO) and its priority for recovery (RTO).
Obviously, there is a cost implication for any backup and with datasets increasing, it can be very expensive to store all your data in multiple, high availability data centers. In some cases, the costs are too prohibitive to justify. Virtualization tools at the server or storage layer often use cloning or snapshot capabilities that serve as ‘back-up’, but these consume space in your production storage which is likely to be the most expensive in the environment
Using one method for priority data backup and another for less important data can reduce costs here. Ideally, mixing disk, tape and cloud storage strikes the right balance between cost and speed. Archived data could sit happily on cheaper tape, but your essential systems, applications, and databases should really be committed to replicated disk. That way, you’ll be ready to restore essential systems rapidly if disaster strikes.
Protecting backups effectively
But it’s not just the process of backing up your data that’s important; it’s what happens to it after. Historically DR processes would have been slow. As datasets have grown, the emphasis has been more on cloud backup rather than disks. Here it’s prone to the same risk of cyber-attack – meaning someone could get hold of your backup as well as the company data, challenging a full recovery. So how do you ensure this data is safe?
There is no one-size-fits-all when it comes to data backup. Whether on the cloud, disk or tapes it’s critical to protect these backups as you would any other data. If using a physical backup, consider storing these offsite in another location, or at least a different building. You may have to qualify this to regulatory audits or your own security assessment. A fire or natural disaster could be all it takes to wipe out all your data along with your backups.
If storing digitally, use a separate file system or cloud storage service that’s located either on a physically or logically separated network. Minimize who has access to these login credentials and keep them on a separate enterprise directory to minimize cyber-attack induced risks. Keeping your data offline and inaccessible s also an effective way of keeping your data out of the hands of cybercriminals. This is known as an ‘air gap’.
Test, test and test again
Backup is an insurance policy for your businesses, but unfortunately, the process is often run on a shoestring budget and deprioritized over other more visible projects – until it’s needed. Most businesses don’t have a backup strategy and if they do, the error is that they’re not testing it. Cyber security frameworks strongly advise regular testing including who to tell if there is a loss, where the backups are stored, how long it’s going to take to recover and how to ensure these backups are stored safely. Automation technology can also locate new servers and applications that have been added to the network and provide notifications if it doesn’t look like it’s been backed up.
Get to your backups in minutes
As data sets get bigger and bigger so recovering 100s of terabytes of data can be critical to your business continuity. You can’t afford for that recovery to take days or weeks. Brands lose customers instantly when a data breach is reported in the media and it can take months or years to undo the damage. But retrieving and securing your data within minutes of a breach or physical disaster goes a long way to reducing the negative impact of data loss.
By knowing your data, what is critical to your business, how to back that data up securely and testing your DR processes regularly, you are much less likely to fall victim to disaster. Proper governance of critical data can maximize revenue, customer satisfaction, and operational cost-efficiency leaving your business resilient against the threat of data loss.
By Ian Richardson, Head of Innovation