There are already many vivid examples on the Internet to explain performance testing, Capacity Testing, load testing, and stress testing concepts, relationships and differences between them, as well as backdoors and other things... here, follow your own understanding to re-statement.
In fact, everything is just the interaction of several factors, and they are functions of each other:
- Concurrency/data volume
- Machine configuration
- Processing speed of a single request
- Stable Running Time
A:
Calculate the processing speed of a single request based on the concurrency/data volume, machine configuration, and the required stable running time (example: all requests can be completed within 2 seconds)
B:
Given the concurrency/data volume, the required stable running time, the processing speed of a single request, and the minimum machine configuration (example: 4 CPU, 16 GB memory is required to support the previous requirements)
C:
The maximum concurrency/data volume (example: supports a maximum of 700 concurrent users per second without loss of stability and better response speed)
D:
Specify the machine configuration, the processing speed of a single request, and the amount of concurrent data, and calculate the stable running time (example: outofmemory 2 hours later)
E:
Combined with C and D, only the configuration of the machine is given, increasing the concurrency/data volume constantly, and observing the response of the other two factors, that is, how the processing speed of a single request changes, and how the stable running time changes
So what are the tests for A, B, C, D, and E respectively?
Xiaoqiang said that bandwidth should also be taken into account. Which test should we consider bandwidth?