What is the difference between bandwidth, throughput, latency and jitter?
Show the answer
Say this first
Bandwidth is capacity: the link rate in bits per second. Throughput is what you actually get end to end, limited by the slowest link, congestion, loss and TCP windows. Goodput is the useful application data after headers and retransmissions. Latency is delay, one way or round trip. Jitter is the variation in delay from packet to packet. Voice and video care most about latency and jitter, bulk transfers about throughput.
Go deeper
Definitions with units
| Term | Unit | Example |
|---|---|---|
| Bandwidth | bit/s | A 1 Gbps port |
| Throughput | bit/s | A file copy reaching 600 Mbps |
| Latency | ms | 35 ms RTT to a cloud region |
| Jitter | ms | Delay varies between 20 ms and 45 ms |
| Loss | % | 0.5% of packets dropped |
Common targets for voice
The usual guidance is one-way delay under 150 ms (ITU-T G.114), jitter under about 30 ms, and loss under 1%. Jitter buffers trade extra delay for smooth playout.
Why throughput falls short of bandwidth
- Header overhead (about 5% for full-size TCP over Ethernet).
- A single TCP flow is capped at window size divided by RTT.
- Loss makes TCP cut its sending rate.
- The bottleneck link anywhere on the path sets the ceiling.
They will ask next
A user has a 1 Gbps link but one download only reaches 50 Mbps to a distant server. Why?
Check RTT and loss first. With 100 ms RTT, a 64 KB window caps one flow at about 5 Mbps, and even with window scaling, small loss rates hold TCP far below link rate on long paths. Multiple streams, a closer server or a better congestion control algorithm help more than a faster link.
Traps
- Using bandwidth and speed interchangeably. More bandwidth does not reduce propagation delay.