Recently after talking with clients we have realised that these terms all appear to mean the same thing to some of the business market out there and this is a dangerous assumption. Typically people only understand the differences after an incident and when it’s too late. Therefore we thought it may be useful to summarise what each is and what it means:
Backup –this typically happens once per day and overnight so therefore it is technically possible to lose around 23 hours’ worth of data. In addition a backup simply backs up data and restore may involve building a new server and then restoring the data to this server which if from tape could mean 24-48 hours before the server is back up and running as was. Therefore in this case the Recovery Point objective (RPO) is 23 hours and Recovery Time Objective (RTO) is 48 hours. With tape this is always assuming each backup is successful and that you have a suitable tape drive to read from.
Business Continuity – This is essentially how your business (as a whole) will recover from an incident. This incident may not affect IT although most incidents do. It maybe related to the scenario that the office is inaccessible due to a gas leak and therefore how do your staff work from another location or it could be a gas explosion within the building which destroys the IT infrastructure. BC is about the capability of the organisation to continue its business following a disruptive incident. BC is the big plan of which IT Disaster Recovery is a subset.
IT Disaster Recovery – is how you plan to recover from a disaster and this isn’t simply restore from backup as let’s not forget you might not have any media to restore from, you might not have an internet link, you may not have servers or tape drives! DR is about how you recover from the worst possible disaster and as quickly as possible. In the business world the only way to guarantee your IT infrastructure to be always available is to host it outside your business in a datacentre (with redundancy / replication) or to host internally but replicate to a remote datacentre using real-time replication, therefore in the event of a disaster your RPO is less than one minute and your RTO is also just minutes. With auto fail-over and the correct replication software this is easily achievable.