Network Time Protocol has been around for over 25 years. In fact it is one of the Internet’s oldest protocols and was in use when the World Wide Web was still in its infancy.
NTP is designed to synchronise devices on a network to the same time. This time can be from any source but most administrators use a UTC (Coordinated Universal Time) time source as that is the global standard for time. This means that networks all around the world can be synchronised to the same time despite being completely separate from each other.
There are various ways of receiving a UTC time source. The Internet is by far the most common but Internet time sources are extremely unreliable and their accuracy relies on their distance away from their clients and the source that they receive their time from. Also Microsoft and Novell warn against using Internet timing sources as a time reference as they can’t be authenticated – a security measure used by NTP to ensure they servers are what they say they are.
A dedicated NTP time server is by far the most accurate and secure method of receiving a UTC timing reference. These devices can receive the signal direct from an atomic clock by a specialist national time and frequency radio transmission that is broadcast by several countries including the US, UK, France, Germany and Japan. Alternatively they can receive the timing signal from the GPS network.
These two methods do vary slightly. The GPS network is the most accurate and also available from anywhere on the planet. the only downside to this is that the GPS antenna needs a clear view of the sky, which can be both impractical and expensive. The radio transmissions, although not broadcast everywhere and vulnerable to interference from local topography can be received by a NTP time server indoors and although not as accurate as the GPS method millisecond accuracy can still be obtained.