Shulou(Shulou.com)06/01 Report--

Performance testing mainly tests the performance of software tests, including load testing, strength testing, database capacity testing, benchmarking, and competitive testing.

1. Load testing

Load testing is a performance test that refers to whether the program can bear the data running in an overloaded environment. In this kind of testing, the test object will take on different workload to evaluate and evaluate the performance behavior of the test object under different workload conditions, as well as the ability to run normally continuously. The goal of load testing is to determine and ensure that the system runs normally even if the maximum expected workload is exceeded. In addition, load testing evaluates performance characteristics, such as response time, transaction rates, and other time-related aspects. For example, the user concurrency test belongs to the load test in the Bhammer S structure. We can use the webload tool to simulate hundreds of customers visiting the website at the same time to see the system response time. What is the processing speed?

two。 Strength test

Strength test is a kind of performance test, which is the performance of the software system when the system resources are very low. This kind of test can often write the software and hardware level requirements of the system. The purpose of implementing and performing such tests is to identify errors caused by insufficient resources or contention for resources. If there is not enough memory or disk space, the test object may show some defects that are not obvious under normal conditions. Other defects may be caused by contention for shared resources, such as database locks or network bandwidth. Strength testing can also be used to determine the maximum amount of work that the test object can handle. For example, a system can run normally under 366m of memory, but it cannot be run under 258m of memory, which tells the system that the memory requirement of this system is 366m.

3. Database capacity test

Database capacity testing refers to inserting a certain amount of data into the database table through a stored procedure to see whether the relevant pages can display the data in time. Database capacity testing allows the test object to process large amounts of data to determine whether it has reached the limit that will cause the software to fail. The capacity test also determines the maximum load or workload that the test object can handle continuously for a given period of time.

For example, if the test object is processing a set of database records to generate a report, the capacity test uses a large test database to verify that the software is working properly and generates the correct report. This test is usually done by writing a stored procedure to insert a certain number of records into a table in the database to calculate the call time of the relevant pages. For example, in an e-commerce system, 10,000 data is inserted into the user table through insert customer to see if it can display the customer information list page normally. If the requirement is met, it can handle up to 100,000 customers, but the customer information list page cannot be displayed within the specified time, so you need to adjust the SQL query statement in the program. If it is displayed within a specified period of time, the number of users can be increased to 20 000, 50 000, 100 000 for testing.

4. Benchmark test

A benchmark is a test that compares with known existing systems and focuses on testing whether it is competitive with similar products. If you want to develop a set of financial system software and you have obtained data such as the performance of Yuyou's financial system, you can test your system to see what is better and worse than Yuyou's financial system? In order to improve their own system, can also provide data for product advertising.

5. Competitive testing

Software competes to use a variety of resources (data records, memory, etc.) to see its ability to compete with other related systems for resources. For example: a machine not only installs your financial system, but also installs UFO financial system. When the CPU share drops, see if it can outperform the UFO financial system, but that your own system can work properly.

According to the statistics of 100 well-known software manufacturers in the United States by the American Center for Software quality and Safety in 2000, it is concluded that the discovery of software defects in the early stage of development is 90% less in manpower than in the later stage of development; the discovery of software defects before introduction to the market is 90% cheaper than that found after launch. So software defects should be found as soon as possible. Not all software has to be tested of any type, and different types of software can be assembled and tested according to the specific conditions of the product.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.