Digging In to Your Data Backup and Recovery Strategy
For your data backup and recovery strategy to be worthwhile, you’ll need to go beyond merely copying your data. Some of the factors you’ll need to delve into include just how well you can recover your data, the operational expenses associated with your strategy, and how well your vendor really supports virtualization.
You copied it, but can you recover it?
Data backups are meaningless if they don’t successfully facilitate data recovery. Tape backups and online backups are two major culprits when it comes to recovery failures.
Tape backups were introduced in the 1950’s and remain in use to this day, partly due to their low cost. But they often offer only an illusion of security. Studies show that anywhere between 50% to 77% per cent of users trying to restore data from tape backups have experienced failures.
Online backups can seem to be a better, more modern solution. The good news is that online backups definitely have a better recovery rate than tape backups. Unfortunately, recovering even small amounts of data from online backups can take a significant amount of time, perhaps even months.
You know what you’d like, but should you spend the money?
In a world where money is no object, businesses would probably demand data backup and recovery plans that allowed for zero data loss and zero time spent to recover data. But in the real world it’s necessary to establish reasonable objectives for acceptable data loss and for the amount of time spent on recovery.
Your Recovery Time Objective (RTO) is the amount of time required to get the crucial aspects of your business back up and running after a disaster. For example, if your RTO for a particular service is zero, that means that the service must be restored immediately. A less crucial service might have an RTO measured in hours or days, depending on how long your business can reasonably function without it. Unsurprisingly, an RTO of zero is going to cost more to support.
Your Recovery Point Objective (RPO) is the point in time to which your data is going to be restored. If your RPO for your data is zero, this means that when service resumes there must be no loss of data at all. This might be necessary for banking or similar applications. In contrast, an RPO of 24 hours or more might be acceptable for some internal reporting applications where the loss of one day of reporting is not significant enough to justify the cost of complete data recovery.
Your vendor says they support virtualization, but what do they really mean?
Webopedia provides this definition: “In computing, virtualization means to create a virtual version of a device or resource, such as a server, storage device, network, or operating system.” This means that, in theory, backing up a virtual machine should involve simply copying its files and configuration data. This copied information should then be easily available to move a virtual machine to different hardware as needed.
However, virtualization can represent a significant load for a host server and for the virtual machines running on that host. For some businesses, having virtual system that operate at a much slower pace will be acceptable. But for situations where speed is necessary, the slower pace may come as an unexpected shock.
Understanding the details of your data backup and recovery plan is crucial to making sure it meets your business needs. You will need to know exactly how recoverable your data really is, how your strategy impacts your costs, and the real facts about how well your vendor supports virtualization.