Preface
At first, I thought that performance testing was just about doing some scripting, parameterization, and correlation, and then throwing out a result after pressing it.
But in fact, it is not only these contents, but also performance analysis. Pay attention to how much response time is improved after tuning, how much TPS is improved, and how much resources are saved.
The direction we strive for is a complete project, which requires early testing, intermediate analysis, and final tuning, not just scripting. If you want to do a good job in performance testing, don't limit your technical scope and cognitive scope.
Whether it is a system, database, code, middleware, storage, or network, whatever problems you encounter, you must try to analyze how to judge them, and consider how to tune them in the subsequent process.
Performance testing is different from functional testing. Functional testing only requires the realization of software functions, while performance testing tests whether the execution efficiency of software functions meets requirements.
For example, if a software has a query function, the functional test only tests whether the query function is implemented, while the performance test requires that the query function is accurate and fast enough. However, for performance testing, it is difficult to define how fast the query speed is fast enough and what kind of query situation is accurate enough. Therefore, some indicators are needed to quantify these data.
Commonly used indicators for performance testing include response time, throughput, number of concurrent users, TPS, etc., which are introduced below.
1.Response time
Response time (ResponseTime) refers to the time required for the system to respond to user requests. This time refers to the time required for the entire process from when the user issues a request to the software client to when the user receives the returned data, including the processing time of various middleware (such as servers, databases, etc.).
The shorter the response time, the faster the software responds and the better the performance. However, the response time needs to be combined with the specific needs of the user. For example, the response time of the train ticket inquiry function can generally be completed within 2 seconds, but when downloading movies from the website, the download speed is already very fast in a few minutes.
The response time of the system will become longer as the number of visits and business volume increases. Generally, during performance testing, in addition to testing whether the normal response time of the system meets the requirements, the response time of the system under certain pressure will also be tested. Variety.
2.Throughput
Throughput refers to the amount of work that the system can complete per unit time. It measures the processing power of the software system server. The measurement unit of throughput can be the number of requests/second, the number of pages/second, the number of visitors/day, the number of business processes/hour, etc.
Throughput is a very important indicator for a software system to measure its own load capacity. The greater the throughput, the more data the system can process per unit time, and the stronger the load capacity of the system.
3. Number of concurrent users
The number of concurrent users refers to the number of users requesting and accessing at the same time. For example, for a certain software, if 100 users request to log in at the same time, the number of concurrent users will be 100. The larger the number of concurrent users, the greater the impact on system performance. A large number of concurrent users may cause slower system response, system instability, etc.
Concurrent access must be considered when designing software systems, and test engineers must also conduct concurrent access testing when performing performance testing.
4.TPS(TransactionperSecond)
TPS refers to the number of transactions and transactions that the system can process per second. It is an important indicator of the system's processing capabilities.
5. Click-through rate (HitsperSecond)
The click-through rate refers to the number of HP requests submitted by the user to the web server per second. This indicator is a unique performance indicator of web applications. The click-through rate can be used to evaluate the load generated by the user and determine whether the system is stable.
The click-through rate is only a reference metric to help measure the performance of the web server.
6. Resource utilization
Resource utilization refers to the software's use of system resources, including CPU utilization, memory utilization, disk utilization, etc. Resource utilization is an important parameter for analyzing software performance bottlenecks.
For example, for a certain software, the expected maximum number of visits is 10,000, but when the number of visits reaches 6,000, the memory utilization has reached 80%, limiting the increase in the number of visits. At this time, it is necessary to consider whether the software has defects such as memory leaks, so as to proceed optimization.
The following is the most comprehensive software testing engineer learning knowledge architecture system diagram in 2023 that I compiled.
1. Python programming from entry to proficiency
2. Practical implementation of interface automation projects
3. Web automation project actual combat
4. Practical implementation of App automation project
5. Resumes of first-tier manufacturers
6. Test and develop DevOps system
7. Commonly used automated testing tools
8. JMeter performance test
9. Summary
As long as you have the ability to do it, you must do it, and don't leave any regrets for yourself. The most important thing in life is not where you stand, but the direction you are heading.
Believe in yourself, firmly believe in your goals, endure hardships and setbacks that ordinary people cannot bear, and keep working hard, and success will eventually be yours!
The fastest and surest way to conquer fear and build confidence is to do what you fear until you gain successful experience.