How to perform stress testing on Web Servers

The company's Web server is ready to go online. How much traffic, response speed, fault tolerance capability, and other performance indicators can it handle? All of these are my biggest concerns. How can we know all this?
 
 
 
You can use tools to effectively test the running status, response time, and other performance indicators of the Web server to solve the above problems. The following uses Web Application Stress Tool (WAST) as an example to describe how to perform Web Stress testing. This is a set of tools specially developed by Microsoft's website testers for actual website stress testing.
 
Test Tool settings
 
Download and install WAST. The process is extremely simple. Before performing a stress test on the target Web server, you must first set it.
 
1. Set the number of concurrent connections
 
Click Defaults → Settings to open the Settings panel. Configure parallel Connections under Concurrent Connections. The Stress Level (Threads) is the least thread, and the Stress Multiplier is the largest thread. Here, the thread specifies the number of threads used by the program in the background for requests, that is, the number of connected clients to be simulated. Generally, it is set to 500 ~ 1000. The number of threads is set based on the local capacity. If you have sufficient confidence in your machine configurations, you can set a higher number.
 
2. Set the duration
 
In "Test Run Time", it is used to specify the duration of a stress Test, which is divided into several unit levels: day, hour, minute, and second. For example, we set it to 1 hour.
 
3. Other settings
 
Use Rpquest Delay to set the Delay time. We can set it to 100 ~ 500. Set the suspension time with Suspend. The Warmup time is the initialization Test Run Time, And the Cooldown time is the test time of the end stage. Bandwith specifies a bandwidth bottleneck. It allows you to simulate the Network bandwidth of a Local Area Network (LAN) connected to T1 (14.4 Mbps) from a 1.5 kbps Modem. Redirects sets the redirection time, Throughput is used to set whether the user, password page status is saved, and Name Resolution is used to set whether to perform Name Resolution. You can set all the above options as needed.
 
Procedure
 
After the configuration is complete, you can perform a stress test. The test procedure is as follows:
 
Step 1: click "New Script" on the toolbar and click "Nanual" in the displayed panel to create a New test project. Set it in the window that appears, and enter the IP address of the Server to be tested in the Server in the main option. Enter 192.168.1.20. Select the Web connection method to be tested at the bottom. Select "get" for "method Verb. Path: select the Path of the Web page to be tested. Enter/Index. asp as the homepage file of the dynamic network. You can set more paths for WAST.
 
Step 2: set the number of Stress Level (Threads) Threads to 1000 in the Settings function. Click the gray triangle button in the tool to perform the test. During the test, we can see from the server's task manager that the CPU usage has reached 100%, and the loss rate has reached the maximum. Run the netstat-an command in the CMD window. You can see that the IP address of the client is connected to port 80 on the server, and the Web site cannot be opened, prompting too many users to connect.
 
Through stress testing, the Administrator has a general understanding of the Web server's ability to withstand stress. The administrator can expand the server hardware based on actual needs and provide a basis for system settings and software selection. It is necessary for a Web server to perform stress testing before it is officially released.