This is the model we set up to measure performance (tps value). The client will send a number of requests from small to large (maximum 3000 requests at a time) to the server. Performance is calculated by the number of requests the server can process in a fixed period of time. Question: Also with this model, in December 2022 we measured tps in the range of 550 ~ 600. However, in December 2023 we measured tps always < 500. Note that EC2 instance type, EC2 OS , server code and client code remain the same. Currently we have not found a way to detect the cause, can you give me any ideas or information to find out what the problem is?
Can find the factor impact to performance value