When disasters strike, there is little that the mankind can do to avert them. Such is nature's wrath. It is purely unpredictable and the magnitude can only be anticipated. In such scenarios, although human lives can be saved through prior intimations, establishments cannot be and thus, the worry for the business organizations. What if a disaster strikes?

Cloud Computing has been smart enough to foresee such instances and has incorporated mitigation schemes to avert such losses. The physical establishments may be hard to recover but what remains of more importance is the data. Therefore, data recovery is the first on the list for the idea is to leap back to normalcy after the disaster withers away. The location of set-up and their disaster vulnerability is certainly a feature of all the cloud computing service providers.

It is very much expected that there will be a certain downtime in case of such occurrences. The first is to classify those risks and their weighted possible damage. There are ideally few kinds of risks as viewed in cloud computing business namely - Natural risks, human induced risks, civil risks, externally factored risks and supplier risks. All these risks refer to categories of events. For instance, an act of terrorism or sabotage will be counted under human risks and the mitigation plan must be well prepared to stock of any such situation

There can be another type which is usually known as facility risks. This type lists risks that are posed by a facility failure - electricity outage, structural or infrastructural flaws, air conditioning failures, fire breakout etc. Ideally, the cloud servers keep a back-up plan in place. The same cluster of data is stored at multiple check points. Therefore, if any one of the infrastructure is lost, the other can be put to function till the previous one is recovered

There are three phases to the disaster recovery plan namely the activation stage, the execution and the final being the reconstitution stage. The first stage addresses the announcement and damage assessment of the disaster; the second included is the execution of the procedures to put the business operations back to the functioning mode and finally the third is the reconstitution of the damaged resources

The activation stage includes three parts. The notification is the first function. The team at the particular resource is responsible to put out an SOS message to the other facilities through any mean of communication. Adding on to this, if possible the team also conveys the damage estimates, response to preliminary recovery measures and most importantly, loss of lives, if any

This would initiate the next level of the stage, i.e. damage assessment. The first part is to take stock of the physical; infrastructure that is left over after the disaster. The next will be the possibilities of recurrence of the same situation and the estimated time to restore the infrastructure. Then, listing of the items to be replaced or repaired is important and then, the final level of the first phase, i.e. planning the activation model

The planning would include the acquisition or requests for the items to be replaced. The communication to the team at other locations is surely a must do and the time estimations with an analysis of the interdependencies of the requisitions ends the first phase of disaster recovery

Once the activation is put in place, the second stage of the disaster recovery takes control of the situation. This may include on-site repairing, site allocation for replacements of the infrastructure. This stage requires informing the users of the resource about the disaster and the time required to pull back services to normalcy. Then acquiring the office and resource supplies are the next on the list. This level also includes the step of restoring as much data or resources possible to put them in a safe custody. Testing system functionality which includes all the security measures comes next and finally, connecting the resource to the other network and other possible resources for speedy restoration. The last level fight is to put together the acquired resources within the estimated time

The last phase of the three stages is that of reconstitution. The first is continuous monitoring of the restored facility. Reassurance of the fact that the re-established facility is fool-proof is very important. After installation of all the hardware, software and the other requisite firmware, implementation of contingency plans is very much a part of the process. The installation process has to be tested few times to assess its operational status. Once the facility is assured operationally, the original staff is intimated about the same to take over the functions and the users are informed of the same

These are the broad steps taken by every disaster recovery team of the cloud service providers. However, different groups might have added stages to their functions but the main concern remains to put the facility back to operation at the earliest possible. Business operations do get affected but the best industry practices have shown that they come to a halt owing to the multiple storage facilities maintained by the cloud computing service providers. But then, God forbid such disasters