Home » Backup Direct Blog » Hosting is not backup – nor is it disaster recovery

Hosting is not backupnor is it disaster recovery

4Posted by Brett Raynes

Two well known hosting providers: Rackspace and Fasthosts have both had complete outages in the past week. We know because we have been contacted by many of their customers who have just realised the consequences...

Both Rackspace and Fasthosts are good companies – Rackspace I know puts a lot of effort into it's service delivery and has reacted with the utmost professionalism and diligence by rectifying the situation and minimising the impact of the problem for its customers.  So this article should not be taken as a criticism.

What the incidents do highlight however – and what their customers have now just realised – is that simply having a computer hosted does not mean it is backed up, nor that there is an effective failover in place.

Most hosting companies offer backup – of sorts. It will normally be made up of some SAN- or tape-based copying process. In other words, your hosted data (let assume this is critical data) is copied on a daily basis to a very large array of disks or a batch of tapes – or even a combination of both.

There are issues with this approach.

Firstly, as a backup it’s a bit crude. It does not give you, the customer, the ability to restore individual files or folders in case they are lost or corrupted – and remember that this type of file loss represents 90% of all losses. All you can do is restore a large chunk or all of the system concerned – something that is so cumbersome single file recovery is not worth doing in most cases.

It’s only really appropriate in a Disaster Recovery situation. And here’s the rub. If there’s a complete outage at the service provider – then access to the backups is probably affected too.  And it’s this situation that arose at Rackspace and Fasthosts. 

One particular soon-to-be-a-customer of Backup Direct, was using Rackspace for both hosting, backup and disaster recover. But they lost everything – the primary servers, the data backups and access to the failover environment. It was a case of having all their eggs in one basket...

I have the greatest respect for Rackspace in particular – but business ought to think twice about relying on a single entity for care of the primary systems and data, as well as their backups. Having a backup and disaster recovery facility with a specialist like Backup Direct means that when the primary host has a problem, they can instantly failover to an independent specialist.

 



COMMENTS

Add your comment

If you have trouble reading the code, click on the code itself to generate a new random code.
Security Code:
 

 
standards