BatchMake is a cross platform tool for batch processing of large amount of data.
BatchMake can process datasets locally or on distributed systems using Condor.


Common Features:

- BSD License
- CMake like scripting langage
- Distributed Scripting via Condor
- Central remote website for online statistical analysis
- User Interface using FLTK
- Cross platform

Application Wrapper:

BatchMake executes precompiled command line applications in a sequential manner.
However, it is often difficult to remember all the command line arguments specifid to an application.
For this reason, BatchMake integrates the Application Wrapper, a simple and intuitive program that allows
users to describe there applications.
Moreover, users who are developing applications using the Insight Toolkit can use the MetaCommand to directly implement
the command line parsing. One powerful option of MetaCommand is that running the compiled application using the
'-vxml' returns an xml description of the current application. The Application Wrapper makes use of this feature so precompiled
applications that uses MetaCommand are parsed directly and do not need to be manually described.

Screenshots of the applications wrapper are also available.
A brief introduction to MetaCommand can be found here.

Distributed Computing:

BatchMake can be run in a distributed environement via Condor. Condor is a high troughput computing system
which allows the use of computation ressources on grids such as Globus.
In a click of a button, BatchMake converts a standard script to a condor script ready to be sent to the Condor manager.
Moreover, BatchMake integrates a novel GUI interface named Condor Watcher to look over the different jobs.
Each job can send data to a central database for data collection.

Online Analysis:

BatchMake is not only a scripting tool, but it also combined a web interface front end and a SQL database for collecting
data from running scripts. An example of such web interface is shown here and some screenshots are also available.
The online interface allow user to define experiments and collect the resulting data from distributed machines and perform
statistical analysis suck as Tagguchi method or graph generation.

Example Scripts:

- MaskCreator.bms
- STAPLEcorrection.bms
- Dashboard.bms (example to send experiment results to the validation dashboard)