Test method for firewall tcp concurrent connections based on maximum new rate

Abstract

A method for testing the number of TCP concurrent connections based on the maximum new rate, which is a method of tearing down and building while including the ramp-up phase, characterized in that: in the ramp-up test, based on the maximum TCP new rate of the device under test, for different The device under test takes a differentiated Ts value, so that the TCP creation rate Vs in the climbing phase is a fixed ratio N of the maximum TCP creation rate Vm of the device; combined with formula 2-1, the Ts value can be obtained through formula 2-3; Ts= Cm / (Vm*N) formula 2‑3; where N can be determined by the testing unit itself. In the present invention (1) the TCP maximum concurrent connection performance test result is less affected by the new TCP performance of the device. In the maintenance phase of this test process, the devices under test work under the same resource load ratio conditions, which can more objectively reflect their respective concurrency performance; (2) It can keep all the devices under test running under the same test conditions and standards.

Description

technical field [0001] The invention relates to a method for testing the number of concurrent TCP connections of a firewall, in particular to a method for testing the number of concurrent TCP connections of a firewall based on the maximum new rate. Background technique [0002] The main goal of the maximum concurrent test is to evaluate the ability of the device under test to manage and maintain TCP connections. Generally speaking, the larger the memory of the device, the more session table space, and the stronger the ability to manage sessions, the higher the TCP concurrency capability of the device. [0003] Generally speaking, there are two test methods for the number of concurrent TCP connections, one is the "build but not tear down" method, and the other is the "build while tearing down" method. The test method of "only building but not tearing down" means that after the TCP connection is established and the specified page is obtained, the connection is kept until the ...

AI Technical Summary

Problems solved by technology

This shows that the proportion of system resources used by firewall B to tear down and create new TCP connections at this stage is much larger than that of firewall A, which also greatly affects the test comparison results of the two concurrency performance

[0014] (2) It is difficult to guarantee that the preset Ts value is applicable to all devices under test

When there are many devices under test and the tests are not performed at the same time, according to the requirements of formula 2-2, the maximum TCP new rate Vm of some devices under test may be less than Vs, resulting in the need to re-determine the value of Ts. Circumstances affecting already completed test items

[0015] From the above analysis, it can be seen that the setting of the ramp time Ts value is an important part of the entire TCP concurrent connection test, and the test method of fixing the Ts value is difficult to solve the above two problems

Images