Performance testing method based on cloud desktop

Abstract

The present invention provides a performance testing method based on cloud desktop, comprising: according to the hardware configuration of the server, estimating the number of virtual machines that the server can carry; running an automated script in the virtual machine to simulate the real scene of the customer; in the server, Put pressure on the estimated number of virtual machines and monitor server performance; adjust the number of virtual machines according to the results of performance indicators to obtain the maximum number of virtual machines that the server can carry. The beneficial effects of the present invention are as follows: in the process of testing the performance of the desktop cloud, there is no need to spend actual manpower and material resources to simulate the office scene, which can be achieved through an automated method; by applying pressure in the virtual machine, the performance index of the server is monitored to achieve purpose of performance testing.

Description

technical field [0001] The invention relates to the technical field of desktop testing, in particular to a method for testing performance of cloud desktops. Background technique [0002] The cloud desktop performance testing technology is mainly to test the number of virtual machines that can be carried by a server with a certain configuration. [0003] At present, although there are similar performance tests in the existing technology, they mainly have the following defects: [0004] 1. Define server performance index standards, which are generally consistent with industry standards. [0005] 2. Failure to evaluate the number of virtual machines on a server with a certain configuration, and not to monitor server performance by increasing or decreasing the number of virtual machines during the stress test to determine the maximum number of loads. [0006] 3. The user simulation operation of the virtual machine is not performed through an automated script, which makes it di...

AI Technical Summary

Problems solved by technology

[0005] 2. Failure to evaluate the number of virtual machines hosted on a server with a certain configuration, and not to monitor server performance by increasing or decreasing the number of virtual machines during the stress test to determine the maximum number of loads

[0006] 3. User simulation operations on virtual machines are not performed through automated scripts, which makes it difficult to simulate a realistic environment

[0007] 4. The server CPU, memory, and disk IO performance data are not monitored to determine whether the performance indicators of the server have reached the bottleneck

[0008] 5. The performance of the server was not monitored during the performance test

Images