BNL, Upton, Long Island, New York, USA
The repository of data for the Relativistic Heavy Ion Collider and associated pre-injector accelerators consists of well over half a petabyte of uncompressed data. By todays standard, this is not a large amount of data. However, a large fraction of that data has never been analyzed and likely contains useful information. We will describe in this paper our efforts to use machine learning techniques to pull out new information from existing data. Our focus has been to look at simple problems, such as associating basic statistics on certain data sets and doing predictive analysis on single array data. The tools we have tested include unsupervised learning using Tensorflow, multimode neural networks, hierarchical temporal memory techniques using NuPic, as well as deep learning techniques using Theano and Keras.
BNL, Upton, Long Island, New York, USA
This paper describes a new web page creation system that allows web developers with limited programming experience to create interactive displays of control system data. Web pages can be created that display live control system data that updates in real-time, as well as data stored within our logging/archiving and database systems. Graphical, tabular, and textual displays are supported as well as standard interaction techniques via buttons, menus and tabs. The developer creates a web page using a custom web page builder. The builder presents a web page as a user-defined grid of tiled cells. The developer chooses the display style of each cell from a list of available cell types, then customizes its data content. Final polish can be applied using HTML and CSS. Specialized tools are available for creating mobile displays. This paper shows examples of the web pages created, and provides a summary of the experience of both the web developers and users.
