It wasn’t that long ago that recovery of a server system required expertise from the infrastructure level right up to the application. In most cases, recovery in the event of a major hardware failure or data corruption involved a number of complex steps, including operating system and application re-installation and finally data restoration. This was a complicated and time consuming process and normally required an infrastructure expert and one or more application experts. You can imagine this was a major issue for a company that had an application server, hosting applications supported by different vendors.
Let’s imagine for instance turning up to the office on Monday morning to discover that the application server hosting the finance and warehouse applications has failed in a major way and the server won’t boot. The onsite IT support troubleshoots and discovers a major hardware component has failed, he immediately calls the hardware vendor. When the hardware vendor arrives around lunch time critical systems have been down for 4 hours and major operations like billing and dispatch have ground to a halt. After replacing a number of components the server is operational again but Windows still won’t boot, as it looks like the data has corrupted on disk. At this point the hardware engineer steps away having done his job, leaving the onsite IT engineer in a world of hurt with operational hardware but corrupted software and critical services still offline. It is at 3pm that the support person bites the bullet and reinstalls Windows and prepares data for restoration, most likely from tape, with also adds time and complexity to the process. Before he can bring the services back online he needs to arrange for reconfiguration and installation of the major applications by the support vendors. If he is lucky he can arrange this for after hours, but most likely it will be next morning before they can reinstall and configure the applications for him. By the time data is restored and services operational, the system will have been down for 36 hours or more.
A major headache in the whole process was requiring external application expertise to re-install and configure applications. These days backup software can snapshot the state of the entire server, meaning that the IT support person doesn’t need to rely on any application expertise to restore the server. In the example above, the onsite IT engineer could have restored a backup of the entire server in a matter of hours once hardware was available. For a small server it is likely it would be back up in under half an hour. Certainly having a virtualised environment simplifies and reduces recovery time, however this type of technology and rapid recovery is also available for physical server installs.
Many companies still employing older data level backup technology are not aware of the complexity and time required to recover from these types of disasters. A company may have a good backup of their data but it doesn’t mean they can automatically recover an entire system in the event of a disaster let alone meet recovery time objectives. Modern backup software and recovery techniques have paved the way for greatly simplified disaster recovery and have reduced recovery time. We have seen this technology trickle down from the enterprise and be available to everyone, even the smallest SMB. Today the process of recovering an entire server is almost child’s play. When I started in IT you needed the involvement of a number of experts if you had any chance at a successful recovery.
To hear the latest improvements in IT disaster recovery and tips on setting up a business continuity plan for your organisation, register for our complimentary business survival guide seminar to be held in Sydney and Melbourne.