EPJ Web Conf.
Volume 245, 202024th International Conference on Computing in High Energy and Nuclear Physics (CHEP 2019)
|Number of page(s)||7|
|Section||1 - Online and Real-time Computing|
|Published online||16 November 2020|
Low Latency, Online Processing of the High-Bandwidth Bunch-By-Bunch Observation Data From the Transverse Feedback System in the LHC
CERN, Geneva, Switzerland
* e-mail: firstname.lastname@example.org
Published online: 16 November 2020
During long shutdown 2 (2019-2020) the transverse observation system (ADTObsBox) in the LHC will undergo a substantial upgrade. The purpose of this upgrade is to allow for true low latency, online processing of the 16 data-streams of transverse bunch-by-bunch, turn-by-turn positional data provided by the transverse feedback system in the LHC (ADT). This system makes both offline and online analysis of the data possible, where the emphasis will lie on online analysis, something that the older generation was not designed to provide. The result of the analysis is made available for accelerator physicists, machine operators, and engineers working with LHC. The new system allows users to capture buffers of various lengths for later analysis just like the older generation and it provides a platform for real-time analysis applications to directly capture the data with minimal latency while also providing a heterogeneous computing platform where the applications can utilize CPUs, GPUs and dedicated FPGAs. The analysis applications include bunch-by-bunch instability analysis and passive bunch-by-bunch tune extraction to name a few. The ADTObsBox system uses commodity server technology in combination with FPGA-based PCIe I/O cards. This paper will cover the design and status of the I/O cards, server, firmware, driver, analysis applications and results of early performance testing.
© The Authors, published by EDP Sciences, 2020
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.