Network latency is the time it takes for data to transfer across a network. It's often called "lag." Networks with a longer delay have "higher latency" while networks with shorter delays have "lower latency." Lower latency is usually preferred because it allows faster communication between devices and services.
Despite their close relationship, latency is not the same thing as network speed. Due to a variety of factors outlined below, high-throughput networks can also experience latency that can impact network performance.
What causes network latency?
In any given network interaction, client devices such as smartphones and laptops communicate with a server on a website or elsewhere using a computer network. For example, an iPhone sends a data request to Amazon's website, which then responds with the requested data.
For the iPhone to communicate with Amazon's servers, the data request travels through several intermediaries before reaching its destination. These intermediaries come in the form of routers (such as eero), network switches, firewalls, data cables, and wireless transmission.
Network latency can occur even on networks with high overall throughput, given that many intermediaries between a local device and the remote server it is communicating with depend on externally-located infrastructure that may be faster or slower than a given local network.
External infrastructure affects latency through:
Transmission medium
Transmission medium refers to the material that data is being transmitted through. A hardwired fiber optic connection has less latency than a wireless connection. Every time the network switches from one medium to another, it adds a few extra milliseconds to overall transmission time.
Travel distance
Longer distances between network endpoints can increase network latency. This is why peer connections to geographically distant servers can take longer than connections to locally located servers.
Data volume
As devices connect to shared data servers (such as the internet), the resulting increase in data volume requests drives a corresponding increase in latency.
Server issues
Most data on the internet is stored remotely on servers in various locations that may be geographically distant from the device or service that is sending a request at any given time. Slow response from the server itself, rather than latency during the network transfer process, can lead to increased perceived latency.