Shulou(Shulou.com)06/02 Report--

The content of this article mainly focuses on how to analyze the principle of nGrinder and compare the performance tools. The content of the article is clear and clear. It is very suitable for beginners to learn and is worth reading. Interested friends can follow the editor to read together. I hope you can get something through this article!

Before introducing nGrinder, it is necessary to say that The Grinder attracts jade.

1. The Grinder

1. The Grinder is an open source performance testing framework based on Java, which is convenient for distributed testing through multiple agent loaders.

2. Main characteristics:

You can test any java code, including a variety of common interfaces

Such as HTTP web servers, SOAP and REST web services, as well as client server RMI, JMS, EJBs, etc., and also supports custom protocols.

The test script uses python and Clojure languages

Mature HTTP protocols support:

Automatically manage client connections and cookies,SSL agents

Support for recording scripts to record and play back complex interactions between browsers and websites.

Including console and Avent end

3. Deficiency

Only one test can be run at a time

No test history

No graphical test report

II. NGrinder

NGrinder is an open source web performance testing platform based on Grinder, which has been redesigned and improved by the development team of NHN, the largest Internet company in South Korea.

Features: open source, easy to use, highly available, extensible

1. NGrinder on the basis of Grinder:

Implement multi-test parallelism

Management based on web

Implement cluster

Built-in svn for easy script editing and management

Support Groovy script, compared to Jython, you can start more virtual users

Realize the monitoring of the target server

Plug-in system extension

2. Working principle

It is composed of a control side controller and multiple agent side agent. The test scenario is established through the control side (browser access), and then distributed to the agent side for stress testing.

Users write test scripts according to certain specifications, and controller distributes the scripts and required resources to agent and executes them with jython.

Collect the running situation, the corresponding time, the running condition of the test target server and so on during the execution of the script. And save these data to generate test reports, which are displayed in the form of dynamic diagrams and data tables. Users can easily see the TPS, CPU and memory of the server under test.

III. Comparison between LoadRunner JMeter and nGrinder

1 、 Loadrunner

Based on UI operation, easy to use. In the early days, it was very popular and powerful, but it was too bulky and troublesome to install.

Do not open the source, the scalability is not high, the charge is expensive. The future direction must be the gradual development of client-side tools to the platform, so it has been gradually replaced.

2 、 JMeter

Based on UI operation, easy to use, but weak programming ability (using beanshell scripting language).

Second, JMeter is thread-based, making it almost impossible to simulate thousands of users.

3 、 nGrinder

A single node can support 4000 to 6000 concurrency, support distribution, monitor the server under test, record scripts, open source, and platform.

The parameterization function is weak

There are high requirements for the tester's code.

Thank you for your reading. I believe you have a certain understanding of "how to analyze nGrinder principles and compare performance tools". Go ahead and practice it. If you want to know more about it, you can follow the website! The editor will continue to bring you better articles!

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.