Research shows that we are more likely to lose our database in the coming year - don't let that happen to you!
Statistical research shows that there’s a high probability we would lose our database in the coming year. It can happen due to all kinds of reasons. Here are just a few:
The disk, on which the database files are located, might fail and stop working. According to this research, there’s a chance of 1.22% this would happen.
Attacks from outside may become routine. We hear all the time about online attacks, and it’s only a matter of time (and luck) until this happens to us as well. The assailant’s purpose is to take over the database, delete it, corrupt it, or hijack it and demand ransom for it.
One of the common reasons for data loss is human error. No way to go around it – we’re all humans, and we all make mistakes sometimes. Problem is that sometimes such mistakes can cost us very dearly. We’ve already seen many cases of a whole table being accidentally deleted, or even the entire database accidentally being dropped or overwritten.
The disks, on which the database is located, are physically inside a server or a storage system. It could be located in an office or in a server farm somewhere. At the end of the day, a catastrophic event outside of our control might happen to said server or storage system. It could be a building fire, a flood, or extreme weather conditions (temperature and moisture). It might sound far fetched and improbable to you, but these things have happened around the world. You don’t want it to happen to you too.
If we agree with the assumption that there’s a high probability we’d lose our database within the next year, then we should definitely do something about it.
This is why the first thing any DBA must do is make sure there are proper periodical backups of all databases under the DBA’s responsibility. But, as it turns out, many organizations don’t backup their databases at all. According to this article, that is about 80% of organizations. Considering the fact that between 40% and 60% of small businesses will not survive a severe data loss, this is, therefore, a very worrying statistic indeed.
But even if you periodically backup your databases – is it enough?
Let’s assume that two years ago someone used some kind of system to perform periodical backups. And let’s assume that once a day this system backs up all the databases into a certain storage disk in the organization’s network. The problem is, no one is monitoring the backup activity, and no one verified it in the past two years. At the moment of truth, when these backups were needed, it turned out that they weren’t working for the past 6 months. Why? Because 6 months ago someone changed the location of the storage disk, or the storage space ran out, or the system performing the backups has failed. There could be a thousand reasons. But no one was aware of it.
Even if the backups work well, many things can happen to these backups. They can become corrupted due to a software or hardware failure, they could have been deleted by mistake… In the end, what’s the worth of all these backups, if we can’t use them to recover the database at the moment of truth? There should be a proper procedure in the organization, that includes periodical tests of full restores. Very few organizations implement such a procedure, to make sure they can actually restore their databases from their backups.
Even if the backups can be restored, in many cases the backup strategy is incorrect or not fitting for the organization’s needs. Every organization has to define for itself how much data they’re willing to lose in the event of a disaster (i.e. RPO – Recovery Point Objective), and how long they’re willing to be in downtime until fully returning to operation (i.e. RTO – Recovery Time Objective). Let’s assume an organization is willing to lose half an hour of data. If we backup the database once a day at midnight, and a critical error happened at 22:00, then the organization has lost 22 hours of data. The business implications of this can be catastrophic.
As part of our database management solution (DBSmart), we’ve got you completely covered in terms of backups.
Analyzing your business needs and requirements (RPO and RTO).
Proper backup strategy planning, giving proper response to the business requirements.
Optimally scheduling the backups in order to reduce workload from the operational system.
Smart management of the backups (storage space, compression, history retention, etc.)
Continuous monitoring of the backups and quick response in the event of a problem.
Backup encryption as per the organization’s needs.
Periodical validation of backups and restore tests.
And most importantly – we’ll be there – 24 hours a day, 7 days a week – to restore your database and bring you back up as fast as possible and with minimal data loss.
Comments