Monday, July 2, 2012

Leap Second Causes Major Hiccups Globally

Who'd have thought that adding one little itty, bitty second  (see previous blog also) would create so much relative consternation and confusion globally? Well, evidently, our current networked computers and  GPS-based system (what will happen when one or more of those satellites ceases operation?) just couldn't handle it. Examples? Only yesterday did we learn that sites such as Reddit, Netflix, Gawker, LinkedIn, Foursquare and Yelp crashed after the leap second was added in order to compensate for the Earth's non-secular change slowdown (see last blog for details).

Reddit announced on Twitter that the added second was causing problems with its Java, and meanwhile Gawker's entire website was down for about 45 minutes. Aggregator, Stumbleupon and review website Yelp also reported problems while harried travellers lost hours as the incremental time change left passengers stranded all across the world.

Up to 50 Qantas flights were delayed by a system glitch with global reservation software 'Amadeus,' which left the airline unable to check in passengers. (Amadeus is the global program that monitors flight weight control, inventory and seat bookings.) This left thousands of passengers from the UK to Australia in turmoil. According to Qantas spokeswoman Lauren Blank :

""Amadeus had a global outage which affects all clients worldwide,"

subsequently conceding the leap second was the reason. Not mentioned, but reported by Qantas passengers were delays created by the inevitable backlog. Tweets from assorted Aussie airports by travelers suffering, reported "big queues and chaos at Adelaide airport as the check-in crashed."

The leap second was added to electronic clocks at midnight universal time on Saturday, with atomic clocks reading 23 hours, 59 minutes and 60 seconds before then moving on to Greenwich Mean Time. (The last leap seconds were added in 2008, 2005 and 1998.) Could the tail end addition (instead of doing it say at midday Greenwich time) have been the cause? Maybe. But without further analysis it's difficult to say exactly what the difference might have been.

What we do know is that any adjustments to atomic clocks are more than a technical curiosity.  A collection of these hyper-accurate devices is used to set co-ordinated universal time (via corrections to the coefiicents 'a' and 'b' - see prior blog), which governs time standards on the internet, satellite navigation, banking computer networks and international air traffic systems. Hence, it is logical that even a tiny change in the adjustment coefficients can wreak havoc if networked systems are overly dependent on them.

Naturally, in the wake of these mishaps, we've heard calls to abandon leap seconds.  But that's easier said than done. What are you going to do instead? Just let the time get more and more out of step with the Earth's rotational timing?  Indeed,  a meeting of the International Telecommunications Union, the UN agency responsible for international communications standards, failed to reach a consensus in January. This is because this issue is complicated and can only be resolved by attention to multiple inputs.

Sure, opponents want a simpler system that avoids the costs (and margins for error)  in making manual changes to thousands of computer networks. But as yet, no one has arrived at any feasible alternative which works as efficiently. Perhaps people just need to get a grip and understand that such an overly intermeshed world of  electronic telecommunications will always be hostage to natural processes that either depart from its expected timing precision (like our slowing Earth) - or can produce effects to disrupt that precision, like CMEs. See, e.g.

I personally think these worriers have bigger 'fish to fry': namely finding a way to get the Repugs to stretch the needed budgets to replace the aging GPS satellites once they cease operation (which may be within the next ten years). THEN, you will see big time disruptions that'll make those caused by the leap second look like a 'sneeze' compared to pneumonic plague.

No comments: