DataForge: modular platform for data storage and analysis

Not scheduled
Oral Information Technology

Speaker

Dr Alexander Nozik (INR RAS, MIPT)

Description

Modern experimental physics is difficult to imagine without computer storage and data analysis. Over the past 30 years, many software packages have been proposed, each of which to some extent solves certain aspects of this global task, but at the moment all of them are more or less obsolete. A common drawback of these systems can be called a "scripting" approach, in which each set of input data is associated with some instruction for its processing. This approach provides the flexibility to work with specific data, but practically closes the possibility of automating the process and organizing parallel and, all the more, distributed computing. There are systems in which scripts control scripts that control scripts, which ultimately leads to the fact that the system becomes very narrowly sharpened for a particular problem and can only be administered by the developer. Solving this problem requires the development of new concepts of scientific software and its implementation on modern platforms.

Primary author

Dr Alexander Nozik (INR RAS, MIPT)

Presentation materials

There are no materials yet.