Whether the disaster is coffee spilled on your keyboard or flight cancellations (and missed meetings or a delayed start to a vacation) due to a massive airline outage, it’s real for you and me. But what if the potential disaster that’s keeping you up at night is much more significant? What if it makes the difference between your company surviving a disaster or becoming a statistic?
A major US airline found itself in the headlines in August when an outage caused the cancellation of 1800 flights and delayed many more. The power outage and subsequent power surge prevented critical systems from switching over to backups causing a domino effect that impacted customer service systems that allow passenger check in for flights, process flight boarding and dispatch aircraft. The total cost of cancelled flights, the vouchers the airline was forced to hand out and hotel rooms for passengers is expected to be in excess of $100M. Ouch.
According to an ITIC Survey in a recent article in the Disaster Recovery Journal, over 95% of large enterprises with more than 1,000 employees say that a single hour of downtime costs them more than $100K per year on average. If the company is an SMB and does not have a datacenter, the risk is even higher – an estimated 25% of those businesses do not come back from an unplanned outage.
A key consideration in disaster recovery planning is geographical disbursement. It is logical to want to position your fail-over systems as close to your operational systems as possible to limit the impact of data transfer latencies and to enable the easy movement of employees, if necessary. However, it is critical to be able to recover data and restore operations in a location that is well outside any potential disaster zone. That means being outside the impact area of a large earthquake or tsunami, not on the same regional power grid, or within the possible path of a major storm.
Unfortunately, the airline failed in this planning, relying solely on backup systems that were located in the same datacenter as the operational systems outside of the local airport.
The distance between datacenters and the latency of the network between them will define the Recovery Point Objective (RPO), or more plainly, the amount of new data at risk of loss. While optical networks are the norm in the Americas, Fibre Channel over IP (FCIP), is a critical tool for regions like Asia Pacific where the high cost of optical networks is prohibitive. Ease of management with FCIP also allows a generalist to do the job rather than having specialists focused on mundane activities.
Hitachi Data Systems VSP family systems combined with the Remote Replication bundle offer a solution for both metro-area synchronous replication with Hitachi True Copy and global-area asynchronous replication with Hitachi Universal Replicator (HUR). The addition of Global Active Device (GAD) provides active-active clustering of 2 storage systems and Hitachi Data Instance Director (HDID) automates and orchestrates it all to meet service level objectives such as RPO and RTO as well as providing additional copy data management functionality.
For more information on Hitachi’s 3DC disaster recovery solutions, click here.
Cisco’s announcement today delivers simplified SAN extension for scaling geographically dispersed datacenters. The new Cisco SAN extension module simplifies data protection and business continuance strategies by enabling backup, remote replication, and other disaster-recovery services over MAN/WAN distances bringing FCIP capabilities to MDS 9700 Series Multilayer Directors.
At Hitachi Data Systems, we’ve partnered strategically with Cisco for over 10 years to deliver BC/DR solutions that make the task of keeping your critical business data safe, secure and accessible as simple as possible.
PS: in case you’re curious about what to do if you spill coffee on your keyboard, IDG answers the question here.