Having your applications in the cloud has many benefits – the main one being that you can quickly and cost effectively scale your IT operation to meet your business requirements. The flexibility of the cloud enables businesses to respond quickly to demand and avoid unexpected costs that can result in budgets being exceeded. Whether it’s public cloud, private cloud or a mixture of both, you can combat security and compliance issues, as well as speed and performance issues by managing data locations.  We’re wisening up to the benefits and uptake continues to accelerate.

But how does moving to the cloud impact on IT resilience? Assuming you can have everything you need from the cloud – you’ve ticked all the boxes and are ready to make the move – either wholly or partly, how can you ensure that your IT availability remains high?

Many businesses seem to think that moving to the cloud will automatically offer them a disaster recovery solution, but they need to explore this a bit further. Moving to the cloud means that you’re moving your applications and data into to a data centre somewhere and accessing them over the internet. So, whilst you may not manage your systems directly, your data is still being stored on hardware and software and your IT system is therefore still susceptible to failure – caused by hardware failure, data corruption, viruses, human error, natural disaster and so on.

Now, your cloud provider should ensure there is adequate disaster recovery protection in place to ensure maximum up-time of your systems, and they may have KPI’s around this. Don’t assume, however, that the disaster recovery service will be adequate for your needs. Most cloud providers will offer as standard a backup and build type of disaster recovery – that is where your data is copied/replicated across to a secondary site (data centre), so your data is secure, but in a disaster the data centre will still have a lot of work to do to revert your services to ‘business as usual’. Following on from recovery of their own systems which will take priority, they will have a number of clients to recover and you could be waiting days until they get around to you.

Some cloud providers will offer a more premium disaster recovery service, typically this is insourced using a disaster recovery specialist through a partner agreement and there is a cost to it. This then begs the question – who should provide my disaster recovery and should it be your cloud services provider or a separate provider? Here are the pro’s and con’s for letting your cloud services provider run your disaster recovery.

Pro’s

  • Having it all under one roof means life will be simpler for you – one supplier, one set of contact details and one level of ownership.  There will be no confusion as to who owns the issue and who should be doing what if the data centre (DC) goes down.
  • When disaster strikes, your team will know who to contact and starting the recovery process will be simple. You’ll probably be aware of who the right person is to contact to get the ball rolling quickly.

Con’s

  • If the data centre hosting your systems goes down, your disaster recovery provider won’t just have your IT systems to recover, they’ll need to recover every customer’ system hosted in their data centre. This is very time consuming and you’ll not know which number you are on the list. Downtime can therefore be much longer than your business can cope with and it’s outside of your control.
  • Standard DR services offered by data centres take copies of your data or virtual machines and store them at another DC. In effect this is a backup service (sometimes called virtual replication also). In order to get your business productive again, your machines need to be reconfigured to work in a new virtual environment. This involves changes to IP addresses, DNS records and host files amongst other configuration to make it all work as before. Then your system needs to be tested for errors and to ensure databases have mounted correctly. All this is done at a domain controller, SQL and application/web server level which means 1015 changes are made to each machine before it can function. This approach to disaster recovery is not very fast and not very reliable, so your IT resilience will be poor.

If speed of recovery is important to you then you’ll want to upgrade from the standard disaster recovery service to a better, more dedicated service that will also be more reliable. An independent provider will be able to meet your RTO and RPO objectives and won’t be caught up trying to recover everyone at the same time.

Another option?

Some specialist disaster recovery providers will be happy to work in partnership with your cloud provider, offering you the best of both worlds – a single point of contact for you, and a disaster recovery service that offers better protection of your business against IT downtime. It’s worth finding a disaster recovery provider which meets your needs and asking them if this is an option – if they (and your cloud provider) want your business they should work hard to make it happen.