Sunday, June 7, 2015

High availability doesn't mean backup!

In a typical discussion about cloud based technologies, not only does the topic come up about knowing where in the world the data is, but there's often a discussion about how the information may exist in multiple places. This usually means that if there is a catastrophic failure in one location the other location is able to provide uninterrupted services.

I am surprised by the number of people who are treating this capability as a backup. It's not! This is what we call high availability.  High availability ensures that critical services remain running, usually in advance of 99% of the time.

Today, most systems support some sort of high availability clustering, making use of real-time replication. As data is added, deleted or modified in one location, it is also updated to other cluster members (often in different geographical locations). If one cluster member should fail, then desktop applications or web browser apps should automatically fail over to the other cluster member. However, in my experience this is not necessarily seamless.

So here is what you need to think about. All of those files that you have stored on dropbox, Google Drive or other file storage services, still need to be backed up. Some of the services have what is called a retention policy. This means that if you accidentally delete a file or maybe even an email, you should be up to restore it within a certain time period without requiring the data to be restored from a backup. However, once that retention period has expired, the file is likely to have been deleted for ever.  In some instances, there may be a capability to permanently delete a file right away.

I know of someone who is currently running a NAS (Network Attached Storage) within their small business. The data on the NAS is replicated to another NAS at another location.  The business considered this to suffice as a "backup".  The problem is, if the data is deleted or corrupted in one location, it is likely to be deleted or corrupted at the other location.  No backup, may result in permanent data loss.

I was discussing this with a former work colleague of mine, Phil Atkinson. Phil specialises in data protection systems at EMC.   He was telling me about a solution called Spanning. Spanning provides the ability to act as a third party tool to backup your business data when using the likes of Office 365, Salesforce or Google apps.  This allows you to comply with your corporate backup policies when moving to the cloud. For a lot of organisations, this will mean keeping data for a lifetime.

I'm hoping to get more of an overview about the capabilities of Spanning. I will keep you up-to-date in the near future.

No comments:

Post a Comment

Related Posts Plugin for WordPress, Blogger...