Backup & Disaster Recovery
The terms Backup and Disaster Recovery (DR) are commonly misused and misunderstood when trying to protect your data. As cybersecurity breaches become more common, it is key to understanding the differences of both before deploying either. In basic terms, having a backup strategy means that you have a full copy of your company data typically now stored off-site. If at the time of reading this your business is backing up to tape, then you need to call us urgently! While having a DR policy in place means that you are able to recover your business-critical systems in the event of a cyber attack or physical breach in hardware.
Backups are typically performed daily and are where you copy your company data to a separate resource. This could be to a tape drive on-site or much better, on a separate server within a secure datacentre in a separate location. This should give you a solid set of data in case you need to restore it if there is a problem. Setting up the backup schedule is a fairly straight forward process and typically is done on an overnight basis so there is less impact on the network during the day. There are quite a few different options on the market and here are a few which Primenet will happily deploy and manage for you.
This old school approach usually means an initial full copy of the data is taken and each subsequent day only the blocks of data that have changed are copied over. Often this data is written to tape and this is stored offsite. The key advantage of this traditional backup method is that it is cost-effective as the tape is relatively cheap. The key disadvantages though are the time it takes to restore - you have to restore the original and then all the incremental backup. This method also has its risks, as if the full backup or any of the incremental backups are damaged, then you don’t have a backup.
At Primenet over the years we have seen more than 50% of tape backups fail to restore, and that’s even if they are taken off-site as it is not uncommon for the backup to be forgotten and left sitting in the drive getting overwritten. If you need to get to your data quickly and ensure your business continuity, then this is when a good disaster recovery policy comes into its own.
Cloud backup has rapidly become the most popular way to back up data. Typically this runs from a web-based management platform and is copied to a secure location on the internet.
Synthetic full backup
A synthetic full backup is typically deployed on a cloud backup. This model is similar to the traditional approach, but each incremental backup is automatically added to the previous backup and is verified to ensure it is complete and can be restored. The advantage of this approach is that a fully verified backup is always available and can be restored rapidly. The disadvantage is that it requires computing resources to reconstitute and verify the data, which can drive up the cost.
Agent / Agentless
To back up a system, a file has to be copied from one medium to another. To do this a program must be running that can check what files have to be backed up, check what files have changed and then copy them to the backup location. With an Agent-based solution, the program is running on the system that has to be backed up and with an Agentless solution, the processing is on a central server that has visibility of the file system that is to be backed up. Both of these approaches have advantages and disadvantages.
Primenet can guide you through all of your backup options and deploy the best solution to suit your needs.
Why Disaster Recovery? - Business Continuity!
Disaster Recovery is so much more than just having a backup. It is about having a technical solution in place that will keep the core components of a Business Continuity plan. Although the technical part of the DR is vital, it is only one part of a much bigger plan.
To create a DR plan you must first identify which key systems are required to keep your business running in the event of an incident that takes your normal infrastructure offline. This could be a natural disaster such as a hurricane, flooding, a fire or even malicious behaviour like a cyber attack or Ransomware attack.
Obviously, one solution is to simply duplicate the infrastructure and replicate the data in real-time, and in the event of a disaster just switch over to that version. This approach can be very effective for some threats but it comes with a hefty price tag.
DR solutions can take many forms ranging from restoring systems from backups to a system that automatically takes over when the primary system is lost, which would be appropriate for systems that don’t need to be on-line quickly and when you could afford to lose some data. An automated backup solution is appropriate for organisations such as an e-commerce company where the loss of availability would have a major financial effect on the business and cannot be tolerated.
It’s likely that budget constraints will play a part in your DR planning and compromises will have to be made. Ultimately your chosen DR plan is likely to incorporate a range of technologies addressing particular issues at different price points.
We at Primenet recommend that you have a backup strategy correlating strongly with a well defined and documented written, and printed out, disaster recovery policy. After all, what's the point of having your well thought out Disaster Recovery plan all documented and stored on your shiny/now failed server only to have no access to it.