In Setting up a disaster recovery (DR) site, is both a business and a technological challenge. The biggest problem crops up during the planning stagewhile trying to strike a balance between expectations coming in from recovery-point objectives (RPO) and recovery-time objectives (RTO); from an organisational point of view. Investment is another factor that plays a critical role in DR site decisions.
A DR sites criticality, as perceived by an organisation for its IT infrastructure, is varied for different industry segments. Thus, justification of a DR investment becomes an issue. DR is like an insurance cover against a calamity. Its often difficult to convince businesses to spend on a perceived calamity. Also, the DR coverage is normally restricted to IT infrastructure alone. Most organisations dont consider DR for other componentsHR or business continuity planning.
Capacity invested in DR IT infrastructure stays idle, till disaster strikes. Thus, the capacity planning for DR becomes a challenge, too. Example; if an organisation decides to keep 50% capacity for IT infrastructure at the DR site, and in case of actual disaster, it gets difficult to scale it up according to business requirements. Hence, careful planning (keeping in mind these myriad dimensions) is an essential component of DR planning.
Once planned properly, involving business decision-makers, then implementation becomes easy, as there are options available in the market. Complete involvement of management, to define the set of expectation, vis-a-vis investment in DR, is an essential part to avoid mismatch of expectationsafter disaster strikes.
Nowadays, there are enough options available from cloud vendors, either in public or private clouds, for DR planning with complete outsourced managed services for its maintenance, support, drills, and switch over. In case of a disaster, a call needs to be made regarding using the services based on infrastructure requirements, criticality of applications (being hosted out of a DR site) and the security risk appetite of an organisation (as per compliance requirement). Nowadays, DR on the cloud is becoming a big thing, a great option.
But, bear in mind that for different industries, compliance requirements for security risks may vary. Based on these parameters, decide the use of private or public cloud services. Normally, private cloud building requires more investment than public cloud. But, security requirements are better met in the private cloud set up. There are limited OS platforms available so far in public cloud. Hence if your application requires specific OS, or specific vendor infrastructure, then public cloud services might not suffice.
In my opinion, if cloud vendors are able to consolidate requirements from customers (having a primary data centre based out of three to four cities and requiring common IT infrastructure DR), then combining all requirements and applying law of probabilityall cities wont be disaster struck, simultaneouslyinvestment can be appropriately sized, to gain scale of economies due to sharing. Cost considerations for such a DR site wont be too high, either.
Thus, DR over the cloud does make much sense, and I believe it will take off in a big way in the days to come. In the end, remember that whatever might be the considerations, DR is not a privilege, but a necessity. The day the disaster strikes the investment will pay for itself.
Add new comment