TY - JOUR
T1 - A GPU offloading mechanism for LHCb
AU - Badalov, Alexey
AU - Perez, Daniel Hugo Campora
AU - Zvyagin, Alexander
AU - Neufeld, Niko
AU - Cardona, Xavier Vilasis
PY - 2014
Y1 - 2014
N2 - The current computational infrastructure at LHCb is designed for sequential execution. It is possible to make use of modern multi-core machines by using multi-threaded algorithms and running multiple instances in parallel, but there is no way to make efficient use of specialized massively parallel hardware, such as graphical processing units and Intel Xeon/Phi. We extend the current infrastructure with an out-of-process computational server able to gather data from multiple instances and process them in large batches.
AB - The current computational infrastructure at LHCb is designed for sequential execution. It is possible to make use of modern multi-core machines by using multi-threaded algorithms and running multiple instances in parallel, but there is no way to make efficient use of specialized massively parallel hardware, such as graphical processing units and Intel Xeon/Phi. We extend the current infrastructure with an out-of-process computational server able to gather data from multiple instances and process them in large batches.
UR - http://www.scopus.com/inward/record.url?scp=84903639277&partnerID=8YFLogxK
U2 - 10.1088/1742-6596/513/5/052004
DO - 10.1088/1742-6596/513/5/052004
M3 - Conference article
AN - SCOPUS:84903639277
SN - 1742-6588
VL - 513
JO - Journal of Physics: Conference Series
JF - Journal of Physics: Conference Series
IS - TRACK 5
M1 - 052004
T2 - 20th International Conference on Computing in High Energy and Nuclear Physics, CHEP 2013
Y2 - 14 October 2013 through 18 October 2013
ER -