It is said that, in life, death and taxes are the only certain things but when it comes to IT systems, failure is guaranteed. Disaster recovery is not about IF systems fail but rather WHEN they do.
Failure arises from a variety of sources ranging from the fact that we are human(to err is human) to sources beyond our control such as natural disasters. People make mistakes; your Database Admin can accidentally drop a database table or your IT staff may forget to buy a crucial support license in order to save costs. Unreliable power necessitates investment in generators that are mechanical devices which are prone to failure and batteries that are a fire hazard. Hard disks fail easily. Floodings and fires are real hazards.
GA Insurance Kenya ran a clever ad campaign that captures this best: it depicted the various risk scenarios ranging from theft to personal injury, all aptly captioned with “@#IT HAPPENS”
Thus, disaster preparedness is not optional today given that businesses are heavily reliant on IT systems and their failure has an impact on service delivery and the bottom line. Disaster recovery plans are developed as part of the business continuity planning process. Critical IT systems are identified along with their Recovery Time Objective(RTO) and Recovery Point Objective(RPO). RTO is the maximum duration of time a service can be unavailable while RPO is the amount of data loss that can be withstood.
With RTO + RPO identified, various strategies are then available to handle or mitigate failure; these can broadly be classified into backups, replication and private cloud. Backups are the simplest of the three and the golden rule is 3-2-1 i.e. keep 3 copies of data on 2 different media and 1 offsite. Replication involves mirroring of data, and private cloud involves replicating the entire infrastructure onto a separate hot or cold site.
Budget availability dictates the choice of strategy, with zero RTO + RPO requiring the most cost. With traditional disaster recovery solutions, businesses have to figure out what their budget can support and make tough decisions about which applications will not be covered.
However, with the cloud and managed cloud service providers such as Node Africa, it is possible to achieve zero RTO and zero RPO at a fraction of the cost. Businesses can lease capacity in the cloud and no longer have to make capital expenditures to acquire redundant infrastructure for their systems. Disaster Recovery as a Service (DRaaS) is about leveraging the cloud for your DR, moving it from the CAPEX to OPEX model.