As the size, sophistication, wealth, and importance of an enterprise increases over time, business continuity through Disaster Recovery (DR) requirements shifts from essentially not providing much protection to providing high-availability access. A small company with only a few customers and a relatively small amount of data can recover from a disaster by taking some minimal, inexpensive precautions that are quick and easy to implement. However, disasters that cause disruption to banks, stock exchanges, utilities, or large companies that rely completely on the Internet for their business can be severely impaired if significant precautions aren’t implemented in advance.
In recent years, business continuity requirements have changed for many enterprises, particularly in response to the threat of terrorism and major natural disasters. What has evolved is the notion of a regional disaster, whereby local, synchronously replicated centers are in regions. However, a regional disaster, such as a major flood or prolonged and widespread power disruption much like what was experienced in New Jersey and New York last year from heavy rains, could concurrently affect both data centers. The shift toward online/Internet-enabled enterprises has also caused companies to build high-availability and continuously computing architectures to support 24x7 business operations, further increasing the need for business continuity. In response, many companies and organizations have re-evaluated disaster readiness capabilities and are implementing geographically dispersed solutions.
To coincide with these business continuity trends, companies are building data centers and technology infrastructures to meet the business requirement for high-availability access. Large financial institutions running synchronous replication are adding third-site, asynchronous, out-of-region DR centers along with a data replication and management infrastructure to detect and manage potentially catastrophic disaster-in-progress events. Companies with high or continuous availability requirements are building adjacent primary and secondary data centers to gain the benefit of synchronous data replication with minimal performance impact. The need for redundant power connections and physical firewall separation for compute and/or storage infrastructure is causing some companies to modify their existing data center infrastructure. The result is the ability to support very high-speed, high-bandwidth networks to support processor-to-processor interconnections for the IBM Parallel Sysplex Coupling Facility, satisfying the business continuity needs of most businesses.
In response to the need for business continuity and high-availability access, many enterprises have re-evaluated disaster readiness capabilities and are implementing geographically dispersed solutions with disaster monitoring and automation. Outages and disasters can occur at any time, and response must be quick but also correct to avoid additional problems. For these reasons, enterprises are analyzing and implementing geographically dispersed disaster software tools that continuously monitor for different types of disaster events, and execute the appropriate scripts to facilitate business continuity based on the type of incident that has occurred, such as a major data center incident or an unplanned link outage. Scripts are automatically tailored by an expert system when invoked based on the configuration and event being managed. By using automatic disaster event detection and the execution of the proper scripts to address the type of recovery needed, human error can be removed from the process, permitting it to be completed as quickly as possible.
Geographically dispersed systems contain a certain amount of built-in intelligence to ensure business continuity in the event one or more geographically dispersed computing centers is lost. Typically, geographically dispersed disaster software tools are constantly communicating between each other at two or three centers, constantly monitoring events for indication of a disaster of some type. In addition, these software tools can determine if a complete computing center has been lost and can take appropriate action to continue providing business continuity.
Today the need for many large companies to have multiple, geographically dispersed computing centers is critical to their continued success should a disaster strike. By incorporating geographically dispersed disaster software tools in each computing center, companies will ensure themselves of business continuity and high-availability.