subject: Disaster Recovery For Servers And Desktops [print this page] Disasters never come with a prior announcement. They always come suddenly and catch you unprepared. If you allow the same thing to happen to your desktop and server, you will be in deep trouble. Consider you have been working on a very important project, have invested a lot of money, time and effort on it and have kept all your data stored in your desktop. What if, your desktop meets a tragedy now? How will you tackle it? Now consider you have a big business company that deals with a huge amount of important data daily and prefers to store all of it in a number of servers. What if the servers crash? Can you afford upsetting your customers? You have to adopt some recovery techniques.
Desktop disaster recovery can be approached with a number of steps:
Find the cause of the disaster. It might be damaged OS, damaged application, hardware failure, some malicious behavior of viruses & Trojans, data loss or a simple user error. Once you know the reason you will be able to deal with it better.
Always have a handy bootable rescue disk. This will save you at the heat of the moment. If you feel your system is about to collapse, simply insert the disk, rescue the critical data, discard the unrecoverable ones and come out clean without having to depend on the OS or a hard drive.
Always have fresh handy system images. Rescue whatever data possible, insert the image disk, re-image the work and catch up where had left. However, remember this cannot be a permanent solution. It is just a stop-gap.
Document the mopping. After cleaning up the disaster, do not forget to document the details of the cause and solution of the disaster for future reference. Multiple crashing of the same system should indicate the solution employed is not the correct one. This is where you will realize how serious the matter is.
Although most of the times the aforementioned turn of events might seem irreversible and fateful, there is something you can do to avoid such a situation. Yes, here th of prevention is better than cure finds a very useful application.
Disaster Recovery includes two main mechanisms: data backup and transaction log. Whether you have stored your data in a desktop or servers, you have to have replicas as a part of the contingency plan. For desktops you can go for conventional storage devices like CDs, DVDs, tapes and external HDDs or choose the more efficient, reliable and secured option of online backups. Backups can be taken in two ways: are of two types: full and differential. The former involves backup of the entire data every time, demanding huge space and completion time.
The latter involves backup of only the changes made ever since the last full backup and demands less time and space. However you must note that a differential backup independently has no value. Transaction log maintains the atomicity of each transaction, i.e. unless a change is ready to be committed to the database, the intermediate information are stored in a log. The business companies with various servers can also maintain a backup server containing the backups from all the other servers, although this isnt a very efficient approach.
