Applicability of GPGPU for Data Quality Monitoring in Large Scale Systems

The contemporary large scale measuring systems in the real-time environment make extensive use of histogramming
as a tool for data quality monitoring. The processing of large number of data channels requires suitable computing power where graphical processors seem to be well suited, especially where there no information exchange between the channels. Histogramming operations run on the central and graphics processing units are discussed. Results of the
performance measurements, including various configuration of the allocation of the histograms in various parts of the memory of used devices, are presented. The multidevice architecture is also discussed.

Author: Krzysztof Korcyl
Conference: Title