Technical Definition of Latency
In network systems, latency refers to the time delay between a request for data and the receipt of the corresponding response. It is typically measured as Round-Trip Time (RTT) or Time To First Byte (TTFB) in milliseconds. While throughput defines how much data can move over a link per second, latency determines how fast a single query can be resolved.
Latency is influenced by physical factors (like the speed of light in fiber optic cables), network routing hops, server processing delays, and queuing queues. In modern systems, keeping latency low (under 100ms) is crucial for real-time applications and SEO rankings.
Primary Causes of Network Latency
- Propagation Delay: The physical travel time of light/data over distance.
- Transmission Delay: The time required to push data packets onto the communication medium.
- Processing Delay: The time the server takes to parse headers and execute business logic.
- Queuing Delay: The time data spends waiting in router or server buffers.