Paul Croteau from Rackspace hosted a very entertaining talk on disaster recovery and the cost of downtime.
Business continuity and disaster recovery are topics that are very close to my heart as I’ve spent a good portion of my time as a consultant in the Middle East and Africa developing, deploying and testing disaster recovery solutions for our customers in the region. I’ve also got to see our solutions get successfully used for real as my region is growing fast, whether from an infrastructure point of view or a political one. This means that whether it’s construction diggers ripping through network cables or riots stopping employees getting to work, disasters do occur.
Two figures were thrown out from the get go:
- 59% of Fortune500 companies suffer at least 1.6 hours of downtime per week – Dunn & Bradstreet
- IT Downtime costs companies $26.5 Billion per year – Information Week
The numbers above are US-centric, however, even though the overall GDP of Middle East, North Africa (MENA) and Sub-Saharan Africa (SSA) are much smaller than that of the USA (according to the World Bank), the average downtimes experienced by the organisations in MENA and SSA are much much higher (due to brittle infrastructure and overall lower investment in business continuity).
These assumptions lead to the conclusion that there is a significant impact to the value of organisations in the Middle East and Africa due to unplanned downtime and disasters.
During the presentation, the information that particularly appealed to me was ways of calculating the cost of downtime. Traditionally, I’ve seen downtime calculated as (revenue / total operational hours) * (number of hours of downtime) as shown in the slide below:
This number can be quite misleading as it does not take into account a whole host of other impacts of downtime. For example, lost sales (or the opportunity cost):
There’s also productivity lost:
As you can see, the value of the downtime increases, and the way the numbers are calculated, you can add them all up to understand the total cost of downtime per hour.
Now here’s where the most important takeaway of the talk was described. We need to take away the emotion from any business continuity discussion and carrying out an exercise to precisely define what the real cost of down would be is the best step in critically analysing disaster recovery requirements.
I have had customer conversations which started out with a whole bunch of emotional thinking and they typically played out in one of two ways:
- I don’t have an accurate value for my business processes and IT services, it sounds like a difficult thing to do and I just need DR protection, so I want to protect everything with a 2 hour recovery time objective (in a couple of cases, protect everything with a 0 hour continuous availability solution).
There is then a difficult realisation when the cost of the solution is calculated.
- I have two or three critical systems that I will use up all of my budget to protect
This usually results in significant outage when a disaster occurs as the two or three critical systems will have survival-critical dependencies that are not met
What I always recommend to customers is that before undertaking any disaster recovery project, take some time and/or budget to get a detailed understanding of what is in your datacentre and how much value individual applications and services have to the organisation.
Even in the case of governmental or not-for-profit organisations, the impact of outages on their services can be quantified and a similar calculation to the one Rackspace showed in their talk can be carried out.