Perfection takes a while to attain, and most of us have to make do with what we have—at least for a while. We believe there are several steps that companies can take to improve their current backup environments and considerably reduce the risks they face in the event of a disaster.
- Build as clear and as detailed a picture of the current backup environment as you can. This is essential in order to understand how long backups take, what the success rate is, if there are problem servers or other bottlenecks, whether the notification and escalation process is in place and is adequate, and whether the executive team are fully briefed about the backup situation. This analysis should also look at which duplicate backups are currently being performed—monitoring access to cloud services like Dropbox and Skydrive is helpful here.
Once this analysis is complete, one can begin to work out where to start. ContinuitySA performs a scaled down basic assessment free of charge for long standing existing clients and can also assist with a very detailed billable Backup Assessment for clients and prospective clients who would like a comprehensive independent analysis of the current health of their backup approach.
- The next step is to turn the backup strategy on its head and re-examine it in terms of recovery. Too many companies look at backups from the point of view how quickly they take place but actually the real metric is how quickly and completely recoverable the data is. In this regard, for example, it’s important to look at whether incremental backups are being performed. These are great for speeding up the backup process but they make recovering from a disaster much more difficult because a whole series of backups has to be catalogued and of high integrity—one failed backup breaks the chain of recovery.
Using your offsite recovery infrastructure, perform a complete recovery test of the entire IT system to understand just how well your current backup setup would perform in the worst-case scenario. As part of this process, companies must understand the full sequence of events involved in a recovery, and how long each one takes (see Figure 1). These include:
- Identification that the loss has occurred and how severe it is.
- Making the recovery decision.
- Getting the right backup tapes back on site and then finding the data.
- The restoration process itself.
The first three steps are often overlooked when rehearsing and timing the recovery process.
- Build up a road map for the recovery process that is aligned with the IT strategy. For example, is the company going to run out of tape capacity for backups given expansion plans or current data growth rates? Is an IT project involving new technology (virtualisation is one good example) going to be affected by the fact that traditional tape backups can’t easily and cheaply accommodate it?
- Finally, it’s necessary to ensure that the recovery times are aligned with the timelines contained in the business continuity plan. Have you been given clear direction from your business heads and executives how quickly systems should be recoverable and how much data you can afford to lose?
Next time, we’ll look at the kind of backup solution that you should be moving towards.