Data Mirroring

Storage Replication

In the event of a disaster, either natural or human-induced, having the information technology (IT) systems that support a business or organization secure is of vital importance. Since the 1970s, when business leaders first began to understand the increasing importance that constant access to data meant to their organizations, interest has developed in providing a safe and reliable means to safeguard data. Beginning in the late 1970s, companies such as Sun Information Systems (later known as SunGard Data Systems) began providing backup computer systems that would preserve a client's data in the event of a disaster. Such endeavors provided what is known as hot site/cold site locations. A hot site is fully equipped to permit an organization to resume operations after a disaster; a cold site does not have that capacity but does preserve data that would otherwise be lost. Organizations that preserved their data at hot or cold sites made a cost-benefit analysis that assisted them in deciding what sort of backup arrangements were needed. These determinations are made as part of the development of a disaster recovery plan.

Disaster recovery plans focus on various control mechanisms that can assist in reducing or eliminating various threats to an organization's operations. Disaster recovery plans are generally a subset of an organization's continuity planning. As such, disaster recovery plans need to include planning for resumption of applications, communications, data, hardware, and other IT infrastructure. In the event of a major crisis or disaster, plans to replace personnel and facilities and to augment communication must also be considered. Strong disaster recovery plans consider three aspects: preventive measures, detective measures, and corrective measures. Preventive measures include those steps that are taken to prevent an event from occurring, such as precluding flood damage or power outages. Detective measures involve using controls to detect or discover unwanted events, such as monitoring systems for computer viruses or hacking. Corrective measures are those that involve restoring the system after a disaster so that operations may resume in a timely manner.

When considering an appropriate disaster recovery plan, organizations must consider appropriate recovery point objectives (RPOs) and recovery time objectives (RTOs). An RPO involves the maximum tolerable period in which data may be lost from an IT service because of a crisis. RPOs provide system operators a guideline for planning for backup services; for example, if the RPO were set at six hours, daily backup would be inappropriate. An RTO represents the duration of time in which a certain process must be restored after a disaster or other interruption of service. RTOs generally include time to fix the problem, the recovery itself, testing, and communication of the problem to users, as well as a return to full service. Data mirroring is a means of providing continual backup of data, thus permitting an RPO of zero lost data.

...