Performance Test Model

 

The performance test model is discussed as follows:
As the traffic per unit time increases, the pressure on the tested system increases, server resources are constantly consumed, and TPS values change due to these factors, and comply with the general rules. The following is a diagram of the performance test pressure change model:
 

Note:
Point A: performance expectations
Point B: higher than expected, system resources at critical points
Point C: higher than expected, and at the inflection point of performance
D: excessive load, insufficient resources, and system crash

Based on the situation shown in the above model, we can roughly divide the current performance test into the following four categories:
1. Performance Test
2. Load Test
3. Stress Test
4. Stability Test

Performance Test
the above model diagram is the criterion. the system performance between point A and point B indicates that the system is put pressure on the system based on the expected performance target, verify whether the system meets the expected performance goals within the range of available resources.
load test
System Performance of point B, it means that the system continues for a period of time under certain pressure until one or more of the system's indicators reach the limit, such as system resource CPU, memory, or IO reach saturation.
stress test
system performance from B to D, it indicates that the system is continuously pressurized when the security load is exceeded until the system no longer accepts the request and can determine a system bottleneck, in order to find out the bottleneck of the system, you need to optimize the system.
stability test
system performance from A to B, indicates that the tested system loads certain business pressure on the system under specific hardware, software, and network conditions, so that the system runs for a long time to check whether the system is stable, generally, the stability test time is N * 12 hours.