Queueing Wait Time Mm1
Calculator

Inputs

Utilization ρ
0.8

Results

Utilization ρ
0.8
Average wait in queue (s)
0.4
Average time in system (s)
0.5
Average queue length
3.2

Computing results

Utilization ρ0.8
Average wait in queue (s)0.4
Average time in system (s)0.5
Average queue length3.2

formula-map diagram

Utilization ρ
0.8
Average wait in queue (s)
0.4
Average time in system (s)
0.5
Average queue length
3.2

Computing relationship

Formula

ρ = λ ÷ μ ; Wq = ρ ÷ (μ − λ)

= 0.8

Note

This is a simplified model: it applies the standard computing formula to the numbers you entered and ignores protocol overhead, compression variability, retries, contention and other real-world effects. Size your systems with measured data.

Example: with the default inputs, calculator.technology.utilization_rho is 0.8.

More in Technology and computing

See all →

Frequently asked questions

What does the M/M/1 queue model calculate?+

It models a single server fed by arrivals at rate λ and serving at rate μ, both in requests per second. The calculator returns utilization ρ = λ ÷ μ and average wait in queue Wq = ρ ÷ (μ − λ).

Why must the service rate exceed the arrival rate?+

If λ is equal to or greater than μ, the formulas divide by zero or return negative values, so the results are meaningless. In the M/M/1 model the queue only has a steady state when μ is greater than λ.

What is the difference between wait in queue and time in system?+

Wait in queue Wq is the time before service starts. Time in system is W = 1 ÷ (μ − λ), which includes both the wait and the service itself.

How is the average queue length computed?+

It is λ × Wq, the arrival rate multiplied by the average wait in queue. This is the average number of requests waiting, not counting the one being served.

Does this model apply to any real system?+

Only as an approximation. M/M/1 assumes one server and random arrivals and service times, so it is an estimate. Real systems with several servers or uneven traffic can behave differently.