Netwerklatency

Netwerklatency is the time required for a data packet to travel from a sender to a receiver and be acted upon. In practice, latency is often discussed as one-way delay, while round-trip time (RTT) measures the time for a packet to reach the destination and for a response to return. Latency is a distinct performance characteristic from bandwidth, which describes the maximum data transfer rate of a link.

Latency results from several contributing delays: propagation delay (the time for a signal to traverse a physical distance), transmission delay (the time to push all packet bits onto the link), processing delay (time spent by devices to examine and forward the packet), and queuing delay (time spent waiting in buffers during congestion). The total latency along a path is the sum of these delays across all intermediate devices and links.

Measurement of latency commonly uses tools such as ping and traceroute, which estimate RTT and one-way delay, respectively. Latency can vary due to network congestion, routing changes, and transient failures, with jitter describing the variability of delays over time.

Typical latency ranges vary by context: local area networks and data centers can have sub-millisecond to a few milliseconds of latency; long-haul internet paths often yield tens to a few hundred milliseconds RTT, with mobile networks showing greater variability. High latency impacts interactive applications, such as online gaming and real-time collaboration, and can influence TCP throughput due to protocol dynamics.

To reduce latency, strategies include optimizing routing paths, reducing the number of hops, deploying edge computing and content delivery networks, prioritizing time-sensitive traffic, and tuning protocols and servers for faster processing. Understanding and managing netwerklatency is essential for applications that require immediate responsiveness or precise timing.