RT Journal Article
T1 Surfing the optimization space of a multiple-GPU parallel implementation of a X-ray tomography reconstruction algorithm
A1 García Blas, Francisco Javier
A1 Abella García, Mónica
A1 Isaila, Florin Daniel
A1 Carretero Pérez, Jesús
A1 Desco Menéndez, Manuel
AB The increasing popularity of massively parallel architectures based on accelerators have opened up the possibility of significantly improving the performance of X-ray computed tomography (CT) applications towards achieving real-time imaging. However, achieving this goal is a challenging process, as most CT applications have not been designed for exploiting the amount of parallelism existing in these architectures. In this paper we present the massively parallel implementation and optimization of Mangoose(++), a CT application for reconstructing 3D volumes from 20 images collected by scanners based on cone-beam geometry. The main contribution of this paper are the following. First, we develop a modular application design that allows to exploit the functional parallelism inside the application and to facilitate the parallelization of individual application phases. Second, we identify a set of optimizations that can be applied individually and in combination for optimally deploying the application on a massively parallel multi-GPU system. Third, we present a study of surfing the optimization space of the modularized application and demonstrate that a significant benefit can be obtained from employing the adequate combination of application optimizations. (C) 2014 Elsevier Inc. All rights reserved.
PB Elsevier
SN 0164-1212
YR 2014
FD 2014-09-01
LK https://hdl.handle.net/10016/35003
UL https://hdl.handle.net/10016/35003
LA eng
NO This work was partially funded by the Spanish Ministry of Science and Technology under the grant TIN2010-16497, the AMIT project (CEN-20101014) from the CDTI-CENIT program, RECAVA-RETIC Network (RD07/0014/2009), projects TEC2010-21619-C04-01, TEC2011-28972-C02-01, and PI11/00616 from the Spanish Ministerio de Ciencia e Innovacion, ARTEMIS program (S2009/DPI-1802), from the Comunidad de Madrid.
DS e-Archivo
RD 18 jul. 2024