Little's law
A rule from queueing theory: average lead time equals average work in progress divided by average throughput.
In more detail
Eighteen items in progress finishing at six per week gives a three-week lead time. Adding people raises both work in progress and throughput and leaves lead time unchanged; halving work in progress halves it. The lever is finishing, not starting.
Examples
lead time = 18 items / 6 per week = 3.0 weeks1,200 messages draining at 50 a second: a new arrival waits 24 seconds.200 threads at 50 ms serve 4,000 requests a second; at 200 ms the same traffic needs 800.Related terms
- work in progress — Work that has been started but not yet finished.
- lead time — The clock time from when a piece of work starts to when it is finished, including all the time it spent waiting.
- throughput — How much work a system completes per unit of time, such as requests per second.
- queue depth — The number of messages currently waiting in a queue.
- drain rate — How many messages per second consumers remove from a queue.
- concurrency — Having several pieces of work in progress during overlapping periods of time, such as requests being handled at once.
- queueing theory — The mathematics of waiting lines, describing how arrival rates and service times produce delay.
Taught in
- SDLC: From Request To Running Software — DevOps & Site Reliability Engineering
- Queues, Events And Asynchronous Work — Scalable System Design & Distributed Architecture
- Capacity Planning As A Design Input — Scalable System Design & Distributed Architecture