latency

Latency refers to the delay between a cause and its effect, or between a request and the corresponding response. In computing and networking, it is commonly defined as the time it takes for a data packet to travel from source to destination, including any processing delays along the path, and is measured as a time interval.

End-to-end latency is typically the sum of several components: propagation delay (distance divided by signal speed), transmission delay (packet size divided by link capacity), processing delay (time spent by devices like routers or servers to examine and forward a packet), and queuing delay (time spent waiting in buffers). Additional delays can arise from protocol handshakes and application-layer processing.

Latency is measured in units of time, most commonly milliseconds or microseconds. It can be reported as one-way latency or as round-trip time (RTT). Some fluctuations are expected, and jitter describes the variation in latency over time.

Typical values vary by context. Local area networks can exhibit sub-millisecond to a few milliseconds of latency. Wide area networks often range from tens to hundreds of milliseconds, with satellite links commonly exceeding several hundred milliseconds. Data centers aim for very low, usually tens of microseconds. Real-time applications such as online gaming, voice over IP, and interactive virtual reality require low and stable latency to maintain responsiveness.

Mitigation strategies include shortening transmission paths, upgrading link speed, and deploying edge computing or caching to reduce distance to users. Protocol and software optimizations, hardware acceleration, and quality of service mechanisms can also help. Latency is distinct from bandwidth; both affect performance, and jitter can degrade real-time experiences even when average latency is low.