Application Performance Engineering Blog – Shunra Software

The BBC reported recently that British technology was pushed to the extreme as the Southern part of England, including the London metropolitan area, was blanketed by a rare snow storm. In London, bus services were canceled and hundreds of thousands turned to the web to find alternative means of transportation.  Web sites for train and taxi services experienced 800% increase in their normal loads, peaking at 32,000 users per second, obviously performance was seriously compromised. Heavy demand also compromised the delivery on Mobile phone networks, mainly due to SMS traffic. People who decided to stay home to work remotely caused a 20% increase demand for broadband.

A significant side effect of the storm was reported by the technology blog Boingboing; The London Metropolitan police lost most of their surveillance capabilities when their CCTV system was literally buried in the snow.

So what should cities with volatile weather do? The answer is quite obvious, other than preparing the infrastructure to be able to deal with the situation, they must also take steps to prepare procedures, communication means adequate for the projected need and obviously need to test those means to scale on a “snowy day.”  Providers must test their networks and web sites for significant spikes on demand in cases of emergency, inclement weather, strikes or any other abnormality in a delivery pattern.  The only real answer is test, retest, and if you still have any doubts, test it again.