Computerworld

Leap seconds may be terminated: ITU

The computing-unfriendly custom may be stopped
  • Joab Jackson (Unknown Publication)
  • 05 September, 2010 22:00

Sparking a fresh round of debate over an ongoing issue in time-keeping circles, the International Telecommunications Union is considering eliminating leap seconds from the time scale used by most computer systems, Coordinated Universal Time (UTC).

Since their introduction in 1971,leap seconds have proved problematic for at least a few software programs. The leap second added on to the end of 2008, for instance, caused Oracle cluster software to reboot unexpectedly in some cases.

Some computer professionals argue, however, that abolishing the leap second at this point will just cause another set of difficulties. The revision "would cause more trouble than it naively claims to circumvent," wrote programmer Rob Seaman, on the Leap Seconds mailing list.

ITU's Radiocommunication Sector (ITU-R), which oversees UTC, is seeking input through October on a potentially revised definition of UTC that does not include leap seconds, an idea that has been under consideration for the past decade. If this working group approves the recommendation, it could go before the next ITU Radiocommunications Assembly in 2012 for final approval, and be implemented by 2018.

The leap second came about as a way to reconcile the growing difference between how computers and humans keep time.

UTC is defined by an iteration of seconds, which are defined with great precision by atomic clocks. Universal Time, in contrast, measures the day by the time it takes the Earth to do one complete rotation, which can fluctuate slightly due to tidal effects. Since 1971, 24 leap seconds have been added on to UTC in order to reconcile UTC and Universal Time.

In the revised ITU plan, the divergence between UTC and UT will be allowed to grow over the next few hundred years, and could be reconciled by a single leap hour at some point.

Most computer systems use UTC, including all those that rely on the Network Time Protocol (NTP). The problem, researchers note, is that leap seconds aren't handled in any sort of standardised way.

"A lack of inexpensive hardware and clock circuitry to correctly label a leap second has resulted in a variety of imaginatively nonstandard ways to represent UTC or zone time near a leap second, which may cause problems trying to synchronise computer clocks near leap seconds," write researchers David Finkleman, John Seago and Kenneth Seidelmann in the paper reviewing the debate, presented this month at the American Institute of Aeronautics and Astronautics conference held in Toronto.

For instance, Unix systems may adjust to the new time by setting their clocks back a second. This approach may "adversely affect," applications such as databases that record data in precise time intervals, the paper's authors assert.

Not everyone thinks eliminating the leap second is a good idea.

The revision "doesn't resolve the underlying geophysical issue or provide a future standards path to make the inevitable much larger and more intrusive adjustments to civil timekeeping that will be needed," Seaman wrote.

Other critics charged that the ITU has not sufficiently considered the effect this change would have on the software community itself, and noted that many systems have already been reconfigured to work with this quirk of time-keeping.

For instance, NTP itself can accommodate leap seconds by use of a parsable file of leap seconds that can be downloaded from the National Institute of Standards and Technology.