Evaluation of the adoption of a Streaming Data Model for Experimental Nuclear Physics
Abstract
While there have been remarkable advances in microelectronics, computing, and data science over recent decades, the research model of Experimental Nuclear Physics (ENP) has remained relatively unchanged. Groups at several institutions are now investigating a streaming data model of readout that takes advantage of these advances. On the time scale of the Electron Ion Collider, streaming detector readout will not only be a viable alternative but may well be the standard readout mode for many experiments. With the future evolution of the 12 GeV program at Jefferson Lab and the ongoing design of the Electron-Ion Collider in mind, we have an opportunity to investigate how moving to streaming readout can fundamentally improve the research model of ENP. Examples of areas of investigation are: <itemlist type="bullet"><item>Rethinking the way measurements are compared to theory.</item> <item>Rethinking the way experimental data are stored and managed.</item> <item>Considering detector design, detector readout and data analysis in an integrated way to take full advantage of the streaming data model</item></itemlist>. In our presentation, we will summarize the progress of ongoing work at Jefferson Lab to evaluate the streaming readout model and how a move to that model influences data storage and processing.
- Publication:
-
APS April Meeting Abstracts
- Pub Date:
- 2019
- Bibcode:
- 2019APS..APRR17005D