Key computer systems at the London Stock Exchange failed earlier this week, preventing large numbers of customers from executing important transactions on one of the busiest days in recent months. I should have thought that the LSE systems would be more resilient than that but clearly not. Such a costly failure does nothing to support London’s aspiration to be a financial capital.
Today another national computer system has been giving problems: the national train reservation system; according to one operator “it’s been up and down all day.”
Fair enough, not being able to buy train tickets is a minor inconvenience, albeit expensive in customer time. Nevertheless, the two failures are symptomatic of poor system design, by not providing adequate resilience and thereby exposing each system to a single point of failure.
Or more accurately, perhaps, a single point of success?
