Working seamlessly
The definitions of High Availability of IT services and Disaster Recovery have evolved over time to mean different things.
It is obvious, both statistically and logically, that a total failure of a critical server or application for whatever reason is more likely to happen in the normal day-to-day working environment than the occurrence of a disaster that destroys the office. So it would be reasonable to assume that providing protection to critical servers or applications would focus first on the higher, more common risk of HA. Yet in practice, the opposite is true. Most SME’s have formal DR plans for their IT systems (even if it is just the off-site copy of a daily backup) while few have any HA plans.
The purpose of both the HA and the DR plan is to get users working as quickly as possible after a failure, with the minimal loss of data. Yet more often than not, the method of achieving this goal is to focus all the attention and the budget on an electronic copy of the data, leaving the complex process of re-building the IT systems to provide access to that data until after the disaster has occurred.
Therefore the objective of SME’s is:
To keep the users seamlessly connected to a working application and data irrespective of the nature of failure, with very low costs, minimal disruption and minimal risk.
HA and DR have been seen as two different components of protection. HA addresses the day-to-day issues (and the likelihood that critical systems will fail) while DR is a commercial necessity to protect against a much less likely threat of disaster, usually associated with “loss of site”.
What has changed is the level of criticality of a growing number of IT systems means that any downtime is a disaster and therefore a disaster can occur without the “loss of site”.
The result is that a comprehensive HA and DR solution should be one of the same thing. The only question is how it is deployed.
We are relying more and more on computers. Take a SME with 20 people. How many of them are relying on the IT? Probably the vast majority (if just for email). Even a day of the system being down is a costly process (both in terms of lost work and emergency IT support cost.)
It's not just about backing up the data, but that the applications always work seamlessly. I know from our own experience that this can be done cost effectively. Gone are the days when the server goes down and we are left looking at blank screens, wondering how on earth we are going to get our work done.
The only thing I will miss is watching the adrenalin charged IT department working on fixing the problem. I always found it strangely interesting watching them rush around whilst everyone else was sipping a leisurely coffee.
Posted by: Philip Woodgate | 04 February 2006 at 09:07
This is a crucially important topic for SMEs. Worryingly, the London Prepared website suggest that "90% of businesses that lose data from a disaster are forced to shut within 2 years". See:
http://www.londonprepared.gov.uk/business/businesscont/index.htm
...and welcome to the world of blogging Daren!
Posted by: David Terrar | 07 February 2006 at 13:58
A truly frightening statistic the source of the data is from the London Chamber of Commerce. A summary of the data can also be found at http://www.hedgeconnect.com/DisasterRecovery/
Posted by: Philip Woodgate | 07 February 2006 at 18:45