per node capacity
The throughput one server or node sustains while staying inside its latency target, found by a load test.
In more detail
It is the one number in a capacity plan you must measure rather than guess, because every other figure is arithmetic on top of it. In the course code the constants PER_SERVER and PER_NODE hold this measured value.
Example
PER_NODE = 1200 queries a second per node, measured by load test, in the capacity plan.Related terms
- capacity — The most work a component or fleet can handle while still meeting its latency target.
- load testing — Measuring how a system behaves while a chosen amount of traffic is sent to it, to find its latency, throughput and limits.
- capacity planning — Estimating how much computing capacity a system will need and when, so it is there before the load arrives.
Taught in
- Back Of The Envelope Estimation: Traffic And Throughput — Scalable System Design & Distributed Architecture
- Capacity Planning As A Design Input — Scalable System Design & Distributed Architecture
- Worked Design Two: A Global Chat System — Scalable System Design & Distributed Architecture