Most organisations understand that it is critical to have a comprehensive disaster recovery plan to prevent downtime and to facilitate faster recovery from interruptions like power outages and natural disasters.
The key is to put the right disaster recovery plan in place.
“A comprehensive disaster recovery strategy requires careful planning to provide businesses with redundancy, geographic distribution options, and back-up power measures,” says Kevin Wee, Director - Colocation Business Development of Asia Pacific, CenturyLink.
CenturyLink believes organisations should consider these five key aspects of an effective disaster recovery plan:
1. The right partner:
With operations-related issues being the leading cause of data centre failures, Wee believes organisations should look for evidence of well-managed, high-availability data centres when choosing an infrastructure partner.
Well-run legacy data centres can perform better than poorly-run data centres with modern designs.
“Data centre design is key to a strong foundation, but business continuity relies on highly-qualified technicians maintaining best practice operational standards,” he adds.
2. Network protection:
The organisation’s network must be secured with hardware redundancy and advanced routing protocols to ensure network connectivity is maintained in the event of a power outage or a natural disaster.
3. Data protection:
The organisation’s mission-critical data and applications should be protected even if network operations are disrupted. This is often best achieved by replicating data off-site in a data centre.
It is important that organisations choose which data, applications and systems are critical to business processes and prioritise their availability over others. This helps to minimise overall business disruption and reduce loss of profits.
Organisations may only discover a weakness in the plan when disaster hits, so it is important to test its validity. Failover procedures must be accurate, including basic elements such as an up-to-date contact list.
Testing should be scheduled once a year, or more often as changes in technology or staff occur, to ensure the strongest disaster recovery plan possible.