Overview

A computer that distributes accurate clock information to other machines on a computer network is called a time server. Its primary job is to ensure that participating systems share a common notion of time. Consistent time across devices is essential for event ordering, log correlation, security protocols, certificate validation, scheduled jobs, and many distributed applications.

How time servers get their time

Time servers obtain a reference time from highly accurate sources such as an atomic clock, a GPS receiver, or other dedicated reference clocks (often called reference or master clocks). Those physical devices are considered the authoritative sources. The time server reads the reference and then makes that time available to client systems using standardized network protocols. In many setups, especially in large organizations, local servers keep time from a connected reference to reduce latency and dependency on remote public services.

Hierarchy and strata

Networked time systems are commonly organized into a hierarchy described by stratum levels. Stratum is a simple numeric ranking that indicates how far a server is from an authoritative reference:

  • Stratum 0: The physical reference clocks themselves (for example, an atomic clock or GPS receiver). These are not network hosts but the ultimate timing sources.
  • Stratum 1: Servers directly connected to a stratum 0 device. These are the usual "primary" time servers that distribute time to a network.
  • Stratum 2 and below: Servers that synchronize to higher-stratum servers. Each hop away from the reference increases the stratum number; lower values indicate closer proximity to the authoritative clock. In practice, typical enterprise networks use stratum levels up to the mid single digits.

Protocols and operation

The most widely used protocol for time distribution is the Network Time Protocol (NTP), which exchanges timing messages to estimate offset and round-trip delay between peers and adjust clocks accordingly. Simpler variants such as SNTP (Simple NTP) exist for devices with limited resources. Time servers run these protocols with carefully tuned parameters: they poll peers at intervals, compute corrections, and discipline the system clock gradually to avoid sudden jumps. Proper operation requires accounting for network latency, jitter, and asymmetry.

Uses, examples and importance

Accurate network time supports many functions: aligning logs for incident response, ensuring correct timestamps on financial transactions, coordinating distributed databases, enforcing time-based access policies, and expiring cryptographic certificates. Embedded systems, industrial controllers, and IoT devices also rely on network time to maintain coordinated behavior. Public and private time server pools are commonly used to scale access, though organizations that require high assurance often deploy their own reference-connected servers.

Best practices and notable considerations

  • Use multiple independent time sources to detect faults and reduce single points of failure.
  • Prefer local stratum 1 or stratum 2 servers over distant public servers to reduce latency and improve stability.
  • Enable authentication or cryptographic protections where available to mitigate spoofing and tampering.
  • Be aware of special cases such as leap seconds and vendor-specific "leap smear" techniques used to smooth adjustments.

Time servers are a foundational service in modern networks. While the principles are simple—distribute a reliable clock—the practical challenges of network timing require careful configuration, monitoring, and attention to security to keep systems coordinated and trustworthy.

More on computers · Networks and infrastructure · Reference clocks · Time synchronization protocols