The Network Time Protocol - Used in Computer Networks

By description, this is a protocol designed for transfer of time around a computer network. It works through a hierarchy composed of several strata starting with Stratum 0 which provides the external source of time. Stratum 1 gets the accurate time from a hardware clock which can be got through GPS signals as they are the safest and most reliable. This is because of the many satellites that orbit the earth at any one time. The references are not limited to these as radio signals and broadcasts also provide a safe and relatively reliable source of time. The broadcasts vary with locations and are not as centralized as the GPS signals but they all provide related times, if not equal. Some common ones include MSF from the UK, JJY from Japan and WWVB from the USA.

Reliability being the main issue means that there is need for backup reference clocks in the unlikely event of a signal outage. These references provide the time with the same precision as the signals and are able to maintain a holdover for extended periods until the signals are regained. These backups are expensive however and the reasons for getting them should be more than the reasons for foregoing them. If you decide you need them to maintain the time then there are crystal oscillators that come at a good price.

NTP client software is usually provided with most OS. Linux has GNU public licence NTP client while Windows has Windows Time service. The advantage of having NTV servers is that is to make sure that the time-stamped transactions are carried out in the specified order. They are the missing link between network time distribution and the global time scale and should be an integral part of any company or organization.