Throughput

Throughput is the actual amount of useful data transmitted over a communication channel per unit of time, usually measured in bits/sec (e.g., Mbps). It is different from bandwidth (theoretical capacity) and latency (delay): throughput is what you actually achieve in practice and can be reduced by protocol overhead, errors, congestion and hardware limits. In network fundamentals and across Audio/Video, Maker and Web contexts, throughput determines how many audio/video streams you can deliver simultaneously, how fast firmware or sensor data can be transferred to a device, and how quickly web assets or API responses reach clients.

Audio/Video: when sending a live 4K stream, viewers may experience stuttering because network throughput is insufficient — even if the ISP advertises 500 Mbps, TCP overhead, packet loss or server CPU limits can reduce actual throughput. Maker: when uploading a large firmware image or dataset to a Raspberry Pi over Wi‑Fi or doing an OTA update for an ESP32, the throughput determines how long the update takes; a slow or noisy link reduces actual throughput and can force retransmissions. Web: for a busy web application, server and network throughput influence how many concurrent users get smooth pages and media; you can measure throughput with tools like iperf or browser devtools and improve it with compression, CDNs, caching or more efficient protocols (e.g., HTTP/2, QUIC).