Outsourcing Disaster Recovery
Jul 9th, 2007 by Rick Barnard
The traditional mindset around disaster recovery has been that it is too expensive for anyone other than the large enterprise. DR meant a second data center with 2x the number of servers and infrastructure and 2x the management complexity (every change had to be replicated at two locations). We’ve all talked about this here before.
Many large organizations kept DR internal in order to:
- Avoid outsourcing costs: Physical assets were guaranteed to be available in the case of a disaster (assets were shared between customers). However the assets did not belong to the customer, but were sunk operational costs. Long-term and inflexible contracts locked customers into outsourcing relationships for 3-5 years.
- Reduce RTO (Recovery Time Objective) and RPO (Recovery Point Objective): Rebuilding a data center environment from bare metal in an outsourced data center was very time consuming (~days). If low RTO and RPO were required, costs went up.
- Leverage existing infrastructure: Outsourcing did not always provide adequate flexibility for testing options and environments. Customers who had underutilized infrastructure in-house could leverage it, but even in the most flexible cases, two data centers were needed (as well as internal processes, data replication capabilities, etc.).
Data center technologies have emerged to make outsourcing DR a viable alternative for mid- and large-sized enterprises alike. Data center virtualization reduces outsourcing costs, lowers RTO and RPO, and enables more flexibility for the customer.
- Lower costs through better efficiencies and increased scale: Data center virtualization lowers costs to the customer, because the outsourcer is able to do more with less. Infrastructure can be repurposed on the fly so it can be extended across a larger number of customers. This provides more flexibility, higher utilization of DR assets, and greater economies of scale.
- Better RTO and RPO: Resources can be shared more easily between customers because provisioning take minutes, instead of days. Service levels to rebuild an environment (RTO) are on the order of minutes (not days).
- Flexibility to the customer: Since resources can be repurposed, additional flexibility is extended to the customer. Testing can be completed often to ensure application availability.
An organization still needs the internal processes and best practices, which an outsourcing company would certainly have.
On top of the value that data center virtualization brings, outsourcers have best-in-class business processes, 24×7 operations, scaled administrative efficiencies, highly available infrastructure (network and power redundancy), and internal failover procedures. A solid understanding of virtualization best practices enhances their value proposition to the customer.
With data center virtualization, there is no better time to look at outsourcing again for cost-effective disaster recovery.