Table of contents
1. Why do performance tests?
Whether the system can handle the expected user load;
whether the system can handle the number of things required by the business;
the system can handle the user load under unexpected circumstances;
the system can be stable under expected and unexpected circumstances and users can obtain a comfortable experience.
Second, when to do performance testing?
After the function test, the function of the system is relatively stable.
3. Basic concepts of performance testing
1. Concurrent number
(1) The number of concurrent users in a narrow sense: the number of users using the same function (send request) of the system at the same time. (Note: Performance tests are generally concurrency in a narrow sense.)
(2) The number of concurrent users in a broad sense: the number of users that put pressure on the server (sending requests) at the same time. (It can be different functions) Generalized concurrency has the sum of requests.
Note: The number of system users: the number of users who have registered in this system The
number of online users: the number of users who log in to the system at the same time
Example: 1000 system users, 500 users are online, 100 users are opening the webpage and browsing, 200 users are performing query operations, and 100 are performing submit operations.
The number of users that put pressure on the server (the number of concurrent users in the broad sense): 300
The number of concurrent users in the narrow sense: 200 users access the query operation at the same time, and 100 users access the submit operation at the same time.
2. Response time RT
ART (Average Response Time)
is the time it takes for a user to send a request to when the user receives a response.
Response time = human reaction time + network transmission time (back and forth) + server processing time + database response time
3. Transaction response time TRT
TPS (Transaction Per Second) the average number of transactions per second Transaction
: a collection of closely related operations Payment
: Accounting system, banking system, a list of requests sent by the membership system, the time it takes for the payment transaction processor to process a transaction .
Example: Subway ticket inspection machine:
There are only ten machines for ticket inspection at the station, and one machine can enter one person in one second. If the
number of concurrent users is 5, then the TPS is 5,
and the number of concurrent users is 10. If the TPS is 10,
the number of concurrent users is 100. Then TPS is still 10
4. Click-through rate
The number of HTTP requests sent to the server per second
5. Throughput/throughput rate
The amount of information processed by the server per unit time
6. Think Time
Time interval between requests and requests
7. Resource utilization
Server resources occupied by the system during operation,
such as: CPU, memory, hard disk, bandwidth, power consumption, etc.
4. Curve inflection point model
This table shows how response time, throughput, and resource utilization change as the number of concurrent users increases.
