Skip to content

Metanome User Guide

Vincent Schwarzer edited this page Nov 22, 2016 · 6 revisions

Table of Contents

Download Metanome

You can download a Metanome build at the download page. The deployment<version>.zip file contains all Metanome binaries and a Tomcat 8 web server. Extract it to a location of choice.

Please note that if you started Metanome once, you should not copy or rename the deployment<version> folder afterwards. Otherwise, you would loose all your data within Metanome, because due to the use of relative pathes the data is then not correctly referenced any more!

Operate Metanome

Start

To start Metanome, extract the deployment<version>.zip and run either deployment<version>/run.bat if you are using Windows or deployment<version>/run.sh if you are using Linux. Both scripts start a Tomcat web server and automatically deploy Metanome. A console will show you the status of the deployment. When the console outputs "Started", your Metanome instance is running and ready to be used.

Now start a web browser like Firefox, enter "localhost:8080" in the address bar and hit enter. The browser then connects to your Metanome instance and you can start profiling.

When you first start Metanome, there are several example datasets and algorithms. Most of the algorithms are only placeholders and do not provide real functionality. A good idea is to first delete all input sources and algorithms in your Metanome instance. Then, add your own datasets and algorithms. You can find existing profiling algorithms at, for instance, the Metanome home page.

Stop

To stop Metanome, go to the console with the server's log messages and hit Ctrl+C. Then, accept the termination request. All data sources, algorithms and results will be stored in the deployment<version> folder. So, if you start Metanome again later, they are still available.

Resource Management

Algorithms

On the Algorithms tab within Metanome, you can remove and add algorithms. You can choose a new algorithm as soon as you obtained the according Metanome-conforme jar-file and placed it in Metanome's algorithm folder:

deployment<version>/backend/WEB-INF/classes/algorithms/

If the algorithm does not appear in the dropdown list although it was placed in the correct folder, please refresh the web page.

Input Sources

Adding new input sources is similar to adding new algorithms. First, place the input file(s) in Metanome's input folder:

deployment<version>/backend/WEB-INF/classes/inputData/

You can also setup a database and place your data there. Then, go to Metanome's Data Sources tab and use the form at the end of that page to specify a new input source. It is very important to specify the correct advanced options for your dataset, for instance, the used separator character, quote character, or escape character. If you make a mistake here, Metanome cannot parse the source correctly!