Software Performance Test and Reliability Test

Performance test:
1. Software Performance Testing includes three objectives: ① discovering defects; ② performance tuning; ③ capability testing and planning

2. Main Indicators of software performance include: response time, system response time and application Delay Time, throughput, number of concurrent users, and resource utilization.

3. The system response time usually refers to the average response time of all functions of the system or the maximum response time of all functions.

4. For non-concurrent application systems, the throughput and response time are strictly inversely proportional.

5. The throughput cannot be infinitely increased. The key reason is that the resource utilization cannot be infinitely increased.

6. resource utilization can greatly help Performance Tuning

7. During stress testing, the software is usually in the performance drop curve.

8. Designers are the core technical forces in software performance testing.

9. when analyzing the performance degradation curve, we will first divide it into several intervals: the performance flat area, the performance slight drop area, and the performance sharp drop area. Among them, the performance flat area is the normal state of the software, so people often want the longer the range, the better; the performance is slightly reduced

It is the buffer zone where the software is subject to high loads. The longer the interval, the better. The sharp drop in performance zone is not the normal operation interval of the software. The response time of this stage will increase sharply to the user's intolerable level, throughput will drop sharply or even lower than the throughput of a single user, but this interval

It plays a major role in analyzing performance bottlenecks. Generally, the starting point (also called the performance inflection point) of the sharp drop-down area is where the performance bottleneck appears, in this case, we can further analyze the resource utilization to find the cause of the performance bottleneck.

10. Failure Recovery testing refers to a test to verify the system's ability to recover from faults. Many systems have some fault tolerance capabilities and can still be used by users in the case of some faults.

11. There are three protection measures against physical damages: (1) All equipment should be protected from disasters (such as floods, fires, lightning, and war. (2) The main equipment should be kept away from man-made damages, such as computers and printers, and should be placed in a data room with security locks. (3) Main devices must

Shielding electromagnetic interference.

Reliability test:
1. At present, the main software reliability analysis methods include failure mode impact analysis, harsh Degree Analysis, fault tree analysis, event tree analysis, and potential line analysis.

2. Reliability Test: Software Reliability Test under heavy business pressure

3. The software reliability test process consists of five steps: determining the reliability target, defining the software operation section, designing test cases, implementing the reliability test, and analyzing the test results.

4. The purpose of software reliability prediction is to predict the failure and effectiveness of the software when it is running based on the fault situations revealed in software reliability and testing.

5. Purpose of software reliability testing: To predict the reliability of software in actual operation through a controlled software testing process

6. the test case set for reliability testing must fully comply with the definition of the running section.

7. Test Cases Used for Software Reliability Testing must meet the requirements of statement coverage

8. Software Reliability Testing is usually used for software with reliability requirements.

9. The indicator of software reliability is the average failure time.

10. In software reliability testing, software failures are the core concern.

11. Software Reliability Analysis methods include failure mode impact analysis, cool analysis, fault tree analysis, event tree analysis, and potential line analysis.

12. Software reliability has two meanings: ① the probability that the software will not cause system failure within the specified time under the specified conditions. ② The ability to execute the required functions under the specified conditions within the specified time period. It has nothing to do with software operators.