Internet time servers are a key piece of today’s digital infrastructure, especially for companies and organizations that rely on atomic clocks to maintain extremely precise synchronization.
However, a few days ago, the United States National Institute of Standards and Technology (NIST) acknowledged that several of its time servers stopped being perfectly synchronized after a power outage on its campus in Boulder, Colorado.
As they explained, the prolonged outage caused the failure of an atomic clock device at said facilities. As a result, campus-hosted Internet time services were no longer able to provide a completely accurate time reference to external organizations using the NTP protocol.
NTP allows clocks to be synchronized over the Internet using data packets and is essential for sectors such as telecommunications, positioning systems and other critical applications. Although NIST systems have batteries and emergency generators, the situation was more complex than expected.
The campus team turned to alternative energy sources to keep the hydrogen maser clocks operational once the batteries were depleted. However, during a later inspection at least one of the key generators was found to have failedwhich disrupted the main signal distribution chain of the atomic clock system.
This caused the NTP servers to become temporarily out of sync with Coordinated Universal Time (UTC).
Jeff Sherman, head of atomic clock systems at NIST, commented that the lag reached around five millionths of a second. Although it seems like a minimal deviation, we must remember that the NIST system It is usually about 5,000 times more precise than a difference of microsecondsso even small mismatches are significant.
This incident is striking because the NIST-F4 atomic clock is considered one of the most advanced time reference systems in the world, and shows that even the most precise infrastructures can have vulnerabilities to unforeseen events such as prolonged blackouts.