|
areyou prepared for a disaster?
In business a disaster can be anything from loss of Data, Server or Workstation failure, loss of Business etc, in fact there are many reasons for business disasters to happen.
planningfor disaster
"Chance Favours the Prepared Mind", (Louis Pasteur) .
When discussing the topic of disaster, often people will make the hopeful assumption that it couldn’t happen to them or that they can sort it out later. Sadly for a large number of companies this is not true.
first somescary statistics
According to the National Archives and Records Administration, USA: 93% of companies that lost their data centre for 10 days or more due to a disaster filed for bankruptcy within one year of the disaster. 50% of businesses without data management for this same period filed for bankruptcy immediately.
If you are large and organised, you’re in big trouble and if you’re not so large or no so organised you’re in big trouble immediately.
According to the 2001 Cost of downtime survey results, companies said the loss of data threatens the survival of a business within: 40% said 72 hours 21% said 48 hours 15% said 24 hours 24% said less than 24 hours
someperspective
There are always disasters than can befall you, but how big a disaster should you plan for and what is the likelihood of it occurring. You need to get a better understanding of what can go wrong, how likely it is to go wrong, what if anything can be done to avoid it or to cope with it and finally how can you balance all these things into a sensible strategy that allows everyone to sleep easily at night without breaking the bank.
It was General Dwight D. Eisenhower who said, "I have always found that plans are useless, but planning is indispensable." Studies have repeatedly shown that those companies who create business plans, marketing plans, any kind of plans but then put them on a shelf and forget about them do significantly better than those that don’t. It is the planning not the plan that makes the difference, it is the thought and learning needed to create it. For this reason don’t outsource your disaster recovery planning. By all means get in experts and other people to help, but be involved.
brainstorm what disasters can happen, not just IT disasters, anything.
Assign rough estimates and remedies to each item in the list. Estimate how often it can be expected e.g. once every 40 years for a flood, or every 3 years for hard disk failure. Estimate the cost to repair or reverse the problem Estimate the cost of living with the problem, of what has been lost. Describe what steps could be taken before the event to mitigate it, e.g. offsite backups, or identifying rented offices in the area.
Go through the list using this information deciding what can be done after a disaster has occurred to recover from it and what can be done before a disaster to avoid it. Not all disasters will be worth protecting against, and some are almost inevitable.
Not all disasters are IT based, have you thought about: Paper records, Electricity Supply, Personnel Access to premises, New Market Developments, Key Client Loss.
- Make it Complete. No gaps in the system, if you need passwords or special hardware know where to get them from.
- Make it Current or as current as you need, how old is that backup data?
- Make it Fast, how long does it take to download 350GB of data over a broadband connection (hint: a looooong time, think weeks or months) . Do we need a local backup as well?
- Make it Cost-effective, can you really afford to completely replicate your entire office just in case?
rootcauses of DataLoss
When considering data stored on computers the most common causes of Data Loss as recorded by Ontrack Data Recovery (a premier provider of post disaster recovery services) and their customers general perception of those threats:
causes of DataLoss Customer Perception % Actual Findings %
- Hardware or System Problem 78% 56%
- Human Error 11% 26%
- Software Corruption or Program Problem 7% 9%
- Computer Viruses 2% 4%
- Natural Disasters 1% 2%
hardware or systemproblems
As the top cause of data loss as well as service interruption, hardware problems need special attention. If we roughly split this into loss of a system, then any critical system requires attention:
do you have:
- Spare network cables?
- Hard drives?
- Power supplies?
- Network switches and routers?
- A spare Internet connection?
It is no accident that most commercial servers have redundant power supplies and hard drives, nor is it a surprise that these can be swapped without loss of service or data and without stopping the server.
For data loss the top of the pile is without doubt the hard drive. The standard life of a modern hard drive is about 3 years, they may show some signs of dying if you are paying attention, but sometimes they don’t. Fortunately they are cheap, so for data that you care about at all use a RAID array (these are standard in many PCs now). If you have an office full of machines assume you will have hard drives fail on you. Have automatic backups from office machines to the server, and backup the server.
when considering datastored on paper
If you have filing cabinets, how can you disaster proof them? What would happen if you had a fire or a flood or a burglar who scattered them around your office like confetti? Can you reasonably backup your filing cabinets? It’s a sobering thought isn’t it.
when considering datastored in people’s heads.
What if people leave, people with experience or key people. What if they get run over by a bus, or are sick, on holiday, or simply somewhere else. Do you have systems with lots of little rules and exceptions that only the secretary really knows? Isn’t she retiring next year?
to Err is Human
Any steps taken to safeguard the business and especially business data should be as automated as possible, especially recovery. Relying on people to remember to do backups or transfer data or any other task in the disaster recovery plan is error prone and is tedious. It is not a happy time to discover that you have forgotten to perform a backup when you need it, nor is it that rare.
An even more common problem is not testing your recovery system, yes the offsite backup is operating regularly but how do you know it can be restored? Do you know how? So it is vital that the actual disaster scenarios are checked from time to time, just don’t do it on a live system, what would happen if it fails?
don't test disaster recovery solutions on live systems
if you must; assume it will fail and act accordingly
andfinally.....
A disaster recovery plan can be a big task, there is nothing that says it has to be completed in one go, in fact it would probably benefit from work over time. Neither is the disaster recovery plan a fixed plan, from time to time it should be reviewed to make sure it is up to date.
articleauthor
David Hutchinson
Director
( Elegenat Software Solutions)
|