Common criteria for measuring software system performance

A common measure of the performance of a software system:

1. Response Time)
The response time is the time that the user feels that the software system is used for its services.
The response time can be subdivided:
(1) Server Response TimeThis time refers to the time when the server completes the transaction request execution, excluding the response from the client to the server (the request and the communication time consumed on the network ), the server response time can measure the server's processing capability.
(2) Network Response TimeThis is the time it takes for network hardware to transmit transaction requests and end the transaction.
(3) client response timeThis is the time that the client spends in building requests and displaying transaction results.
The response time the customer feels is actually equalClient Response Time + Server Response Time + network response time

 

2. Throughput)
Throughput reflects the "meal" of the software system, that is, the processing capability of the system. Specifically, it refers to the number of transactions, requests, and data units that the software system can process per unit of time.
The throughput definition is flexible and has different interpretations in different scenarios.
For example
The database throughput refers to the number of SQL statements executed per unit time;
The network throughput refers to the data traffic transmitted over the network per unit time.
The throughput is determined by the load (such as the number of users) or behavior.

3. Resource Utilization)
Common resources include CPU usage, memory usage, disk I/O, and network I/O.

4. clicks (hits per second)
The number of clicks is a useful indicator to measure the processing capability of the Web server.
The number of clicks is calculated based on the number of HTTP requests that the client has initiated to the Web server. A single mouse may trigger multiple HTTP requests, this requires computing based on the specific web system implementation.

5. concurrent users)
The number of concurrent users is used to measure the server's concurrent capacity and synchronization and coordination capabilities. A client refers to a batch of Users performing an operation at the same time. The number of concurrency reflects the concurrent processing capability of the software system. Unlike throughput, most of the concurrency is occupied by operating system resources such as sockets and handles.

In addition, the performance indicators of the software system include the system recovery time. In fact, all user requirements on resources and time can be regarded as the measurement of the software system, performance testing is to verify whether these performance indicators are met.