Publications Repository - Helmholtz-Zentrum Dresden-Rossendorf
|Total number to be selected: 1 Title record|
Data processing performance analysis for ultrafast electron beam X-ray CT using parallel processing hardware architectures
Bieberle, A.; Frust, T.; Wagner, M.; Bieberle, M.; Hampel, U.
The ultrafast electron beam X-ray computed tomography measuring system (Fischer et al., 2008) of the Helmholtz-Zentrum Dresden - Rossendorf (HZDR) is primarily operated for fundamental multiphase flow investigations, e.g. in various technical devices, and for validation of enhanced flow simulation models, e.g. developed for computational fluid dynamic codes (CFD). The ultrafast computed tomography (CT) scanner delivers cross-sectional material distributions by contactless measurement with a spatial resolution of approximately 1 mm and a temporal resolution of maximal 8 kHz. Currently, two central time-consuming processes have been identified limiting the efficient usage of that worldwide unique CT technique: a) the data transfer from the detector system to central data storages (e.g. computer or data base) and b) the data processing. Thus, data processing and data reconstruction has been adapted for the use at multi-core central processing units (CPUs) and many-core graphics processing units (GPUs). For optimal performance an advanced performance PC (AP-PC) with two parallel operated high performance graphics processing units (Tesla K20c, NVIDIA®), a six-core Intel® processor (Xeon E5-1650 v3,), a high internal data bus speed and a large memory block (DDR4, 2133 MHz, 128 GByte) was assembled. Finally, the modified combined multi-core CPU and many-core GPU optimized algorithms generate a performance improvement of app. 137 for the entire data processing sequence compared to the established single core CPU based data processing tool.
Keywords: computed tomography; many-core graphics processing units; multi-core central processing units; massive parallel data processing
Flow Measurement and Instrumentation 53(2017), 180-188
Online First (2016) DOI: 10.1016/j.flowmeasinst.2016.04.004