One of the greatest emerging benefits of the cloud is the possibilities it offers in disaster recovery (DR). Some are even calling it cloud’s killer app.
It is not an exaggeration to say that it completely changes disaster recovery, turning what was a costly and time consuming function into a simpler and far cheaper one.
In short, DR on the cloud means easier, faster and cheaper recovery of applications with fewer worries about backups and restoring data.
For organisations that have to have strong disaster recovery safeguards, for example retail or financial services businesses, the cloud offers superior disaster recovery for less cost. For those smaller organisations that have not implemented disaster recovery owing to cost, those options are now affordable.
Despite this, cloud-based disaster recovery is still in its infancy. According to a recent survey by SearchStorage.co.uk, only 11% of respondents were using cloud for disaster recovery.
This will change as cloud-based DR offers low hanging fruit just asking to be plucked by CIOs because of its numerous benefits:
- With the infinite storage cloud offers you can stream your data to the cloud and you don’t need to plan the amount of space required.
- Larger organisations can do away entirely with their disaster recovery data centres and the activities required to maintain them.
- In a worst-case scenario where your premises are wiped out, you will not have to worry about losing your data stores as they are all off site.
So how do you decide what solution is right for your organisation? There are two well established metrics for conventional IT disaster recovery that serve as a useful guide when working out what sort of service level you require from a cloud-based DR solution:
- Recovery target objective (RTO) – if the system went down, how long would it take to recover?
- Recovery point objective (RPO) – if the system went down, how much data would be lost?
Defining these will help decide what option is most appropriate for your organisation.
Cold, warm and hot solutions
The choice of recovery options depends on the available budget and the degree of downtime the organisation is prepared to accept.
Cold Standby: The secondary node acts as backup to another identical primary system. It will be installed and configured only when the primary node breaks down for the first time. Subsequently, in the case of a failure in the primary, the secondary node is powered on and the data restored before finally starting the failed component. Data from the primary system can be backed up on a storage system and restored on the secondary system as and when required. Recovery time is typically a few hours.
Warm Standby: The software component is installed and available on the secondary node. The secondary node is up and running. In the case of a failure on the primary node, these software components are started on the secondary node. This process is usually automated using a cluster manager. Data is regularly mirrored to secondary system using disk based replication or shared disk. Recovery time is typically a few minutes.
Hot Standby: Software components are installed and available on both primary and secondary nodes. The software components on the secondary system are up but will not process data or requests. Data is mirrored in near real time and both systems will have identical data. Data replication is typically done through the software’s capabilities. Expected recovery time is just a few seconds.
Active-Active (Load Balanced): In this method both the primary and secondary systems are active and processing requests in parallel. Data replication happens through software capabilities and would be bi-directional. This generally provides an instantaneous recovery time.
Which option to choose generally comes down to cost, which has always been a key parameter in disaster recovery, but the cloud changes the equation. Whereas you might have had to go with ‘cold’ in the past, with cloud it is now cost effective to deploy a ‘warm’ solution. Further, technologies like Amazon Web Services’ Pilot Light allow for more cost effective solutions compared to traditional DR datacentres.
Cloud is also blurring the lines between DR and backup. With Azure, for example, you can have much improved RTOs and near instantaneous RPOs at a price not much greater than what you are paying for backup. Microsoft Azure Site Recovery replicates near instantaneously into the cloud. If you do have a failure of either data centre or servers you can rapidly restore that service into operation.
Switching disaster recovery to the cloud is not that difficult or complex and organisations that avoid making the switch are missing a trick.
Disaster recovery used to be painfully expensive and resource intensive. Cloud now allows you to dramatically change that. It is a relatively simple change to make which is why so many leading organisations are getting on and doing it.