Can the cloud achieve ‘zero loss’ instant recovery following an IT failure?
Wed 3 Jun 2015
Instant failover with zero data loss is the Holy Grail of maintaining a resilient IT system. With businesses demanding more and more from their IT systems, there is less leniency towards downtime. Customers expect suppliers to have adequate disaster recovery provisions, as the knock-on impact to their business is damaging.
Businesses now don’t have much choice, other than to03 bite the bullet and invest in reducing the risk of IT downtime. There’s a lot of noise around disaster recovery – especially with regards to cloud disaster recovery, but what is really possible in this day and age?
Cloud disaster recovery means that copies of your data are maintained in the cloud. This is beneficial as your data is off-site and maintained by a third party, often at attractive prices. However, it’s the ‘recovery’ element which you need to pay particular attention to, and this is the part that makes the difference between a business that can continue to be productive following an outage, and one where business productivity ceases, with big financial impacts to the company.
The current state of play is that you’ll have to compromise on either speed of recovery or data loss if you opt for cloud disaster recovery. If your IT systems fail, you’ll need to decide would you rather be up and running instantly, whilst losing the last few hours of data, or return to productivity in a few hours, whilst keeping all data up to the moment before the incident?
The future of disaster recovery will see pre-recovery and continuous replication brought together to achieve both instant failover with zero data loss
For most companies experiencing an IT outage, the speed of return to service is a top priority, whether it’s a full or partial loss. But can we have fast recovery with zero data loss? No is the simple answer with current cloud disaster recovery. Recovering IT systems is a complex task. Although most companies now have easily accessible copies of their data, it’s making the systems work together, and the configuration to enable businesses to be fully productive again that takes time.
A cloud disaster recovery provider who is recovering after an incident will achieve recovery in 4-48 hours. Instant recovery, on the other hand, is now possible – but only by performing the recovery process in advance of a failure. Called pre-recovery, a virtual standby solution gives companies the option of recovering instantly from an IT failure. However, in order to recover in advance, a point in time data replication needs to be taken. This means companies may lose a few hours’ worth of data.
Continuous replication is available for companies that want to achieve minimal data loss, but because recovery is not done in advance recovery times are unreliable. If speed of recovery isn’t essential then this may be an option for you.
The future of disaster recovery will see pre-recovery and continuous replication brought together to achieve both instant failover with zero data loss. It’s in the pipeline, but will require more development before it comes to market. Until then it’s a decision between speed of recovery or data loss.