The stability of the server is the most important thing for us. If the performance cannot guarantee the needs of business operation, then it is a waste of money. Today Yijibi
Xiaoxin will explain to you how to test server performance from the following perspectives.
Stability test:
Know the number of users of the system during peak periods, frequency of each transaction operation, etc. Design a comprehensive test scenario. During the test, run each scenario together according to a certain number of people to simulate the
number of users. And monitor whether the performance indicators of the system can maintain normal values under this pressure during the test. Will the transaction response time fluctuate or
increase as the test time increases? Whether the system will have abnormal conditions such as downtime and application suspension during the test period.
pressure test
Knowing the number of users in the peak period of the server system, verify whether the transaction response time of each transaction can meet the customer's requirements under the maximum concurrent number (converted by the number of people in the peak period). Whether the performance indicators of the system
are still within normal values under this pressure. Whether the system will cause adverse reactions (such as downtime, abnormal application termination, etc.) due to such pressure.
fault tolerance testing
By simulating some abnormal situations (such as sudden power failure of the server, intermittent network, insufficient hard disk space of the server, etc.), it is verified whether the system can have an automatic processing mechanism to ensure the normal operation or recovery of the system when these situations occur. measure.
If there is an HA (automatic disaster recovery system), additional tests can be conducted specifically for these automatic protection systems to verify whether they can effectively trigger protection measures.
Troubleshooting test
Based on original cases or experience judgments, verification tests are carried out on modules that have had problems or are suspected to have hidden dangers in the system, and verify whether the same
performance problems will occur in these modules. For example: the memory leak problem of the upload attachment module, the optimization of the address book module, the impact of turning on TIvoli performance monitoring on OA system performance, etc.