Networking

Bandwidth Calculator

Estimate a shared network capacity target from the number of users or devices, their average demand while active, and the share active at once. Add a safety margin to compare a working demand estimate with a capacity target.

Your inputs

The share active at the same time. 50% of 20 devices means 10 active-device equivalents.

Leave blank for 0%. Margin is added to estimated demand.

Calculated on your device. Raw input values are not sent to analytics.

Your estimate

Estimated total bandwidth
50 Mbps
Recommended bandwidth
60 Mbps

This planning model averages concurrent demand. Bursts and applications with guaranteed capacity requirements may need more headroom. A zero usage estimate does not mean a network has no background traffic.

Estimated total bandwidth: 50 Mbps. Recommended bandwidth: 60 Mbps

What bandwidth does your network need?

Bandwidth describes the capacity of a connection. Throughput is the useful rate actually achieved. This tool combines device demand into a planning estimate; it does not predict application latency, measure congestion or guarantee that a particular internet service meets every workload.

Bandwidth formula

Estimated Mbps = device count × Mbps per active device × simultaneous usage percentage ÷ 100. Recommended Mbps = estimated Mbps × (1 + safety margin percentage ÷ 100). Divide Mbps by 1,000 to express Gbps. A Kbps input is divided by 1,000 before calculating.

For 20 users averaging 5 Mbps while active, with 50% simultaneous use, demand is 50 Mbps. A 20% margin gives a 60 Mbps target. Concurrency is an average planning fraction and may produce fractional active-device equivalents; it is not rounded up to a whole device.

Concurrency and bursts

If every device must work at once, use 100% simultaneous usage. A lower value is appropriate only when your workload supports that assumption. Starting many downloads or meetings together can create peaks far above the average. Model the busiest meaningful interval rather than averaging activity across an entire day.

Overhead, margin and throughput

Safety margin is additional capacity above modeled demand. It can cover uncertainty and allow headroom, but it is not a detailed model of protocol overhead or quality of service. Do not both inflate per-device demand and add the same margin without intending to be conservative.

Check upstream and downstream needs separately. Video meetings and cloud backups can need upload capacity even when the download side appears lightly used. Port rates, wireless airtime and other bottlenecks along the path can also constrain throughput.

Mbps versus MB/s

Network rates use bits: 8 bits equal 1 byte. A measured file copy at 10 MB/s represents 80 Mbps before additional overhead. The per-device input accepts Kbps or Mbps, so convert byte-based rates first. For file completion times, use the linked transfer or download calculator.

Examples

Small office

Input
8 users × 3 Mbps, 75% simultaneous usage, 20% margin.
Result
18 Mbps estimated; 21.6 Mbps recommended.

20 users

Input
20 users × 5 Mbps, 50% simultaneous usage, 20% margin.
Result
50 Mbps estimated; 60 Mbps recommended.

50 users

Input
50 users × 8 Mbps, 60% simultaneous usage, 25% margin.
Result
240 Mbps estimated; 300 Mbps recommended.

100 users

Input
100 users × 10 Mbps, 80% simultaneous usage, 30% margin.
Result
800 Mbps estimated; 1,040 Mbps (1.04 Gbps) recommended.

Frequently asked questions

What is simultaneous usage?

It is the share of users or devices expected to demand the entered average rate at the same time. Use 100% when all devices must be served concurrently.

Does the recommended rate guarantee performance?

No. It is an arithmetic capacity target. Real throughput, latency, traffic priorities and individual application requirements still need assessment.

Can I enter Kbps?

Yes. The calculator divides Kbps by 1,000 to obtain Mbps. For example, 500 Kbps per device is 0.5 Mbps.

Why add a safety margin?

A margin leaves capacity above your average demand estimate. It can help accommodate uncertainty and peaks, but the appropriate amount depends on your workload and service requirements.

What does zero simultaneous usage mean?

It produces zero modeled application demand. It does not mean the network has no background traffic or that no network capacity is needed.

Last reviewed: