TY - GEN
T1 - From knights corner to landing
T2 - 32nd International Conference on High Performance Computing, ISC High Performance 2017
AU - Chatzikonstantis, George
AU - Jiménez, Diego
AU - Meneses, Esteban
AU - Strydis, Christos
AU - Sidiropoulos, Harry
AU - Soudris, Dimitrios
N1 - Publisher Copyright:
© Springer International Publishing AG 2017.
PY - 2017
Y1 - 2017
N2 - Brain modeling has been presenting significant challenges to the world of high-performance computing (HPC) over the years. The field of computational neuroscience has been developing a demand for physiologically plausible neuron models, that feature increased complexity and thus, require greater computational power. We explore Intel’s newest generation of Xeon Phi computing platforms, named Knights Landing (KNL), as a way to match the need for processing power and as an upgrade over the previous generation of Xeon Phi models, the Knights Corner (KNC). Our neuron simulator of choice features a Hodgkin-Huxley-based (HH) model which has been ported on both generations of Xeon Phi platforms and aggressively draws on both platforms’ computational assets. The application uses the OpenMP interface for efficient parallelization and the Xeon Phi’s vectorization buffers for Single-Instruction Multiple Data (SIMD) processing. In this study we offer insight into the efficiency with which the application utilizes the assets of the two Xeon Phi generations and we evaluate the merits of utilizing the KNL over its predecessor. In our case, an out-of-the-box transition on Knights Landing, offers on average 2.4x speed up while consuming 48% less energy than KNC.
AB - Brain modeling has been presenting significant challenges to the world of high-performance computing (HPC) over the years. The field of computational neuroscience has been developing a demand for physiologically plausible neuron models, that feature increased complexity and thus, require greater computational power. We explore Intel’s newest generation of Xeon Phi computing platforms, named Knights Landing (KNL), as a way to match the need for processing power and as an upgrade over the previous generation of Xeon Phi models, the Knights Corner (KNC). Our neuron simulator of choice features a Hodgkin-Huxley-based (HH) model which has been ported on both generations of Xeon Phi platforms and aggressively draws on both platforms’ computational assets. The application uses the OpenMP interface for efficient parallelization and the Xeon Phi’s vectorization buffers for Single-Instruction Multiple Data (SIMD) processing. In this study we offer insight into the efficiency with which the application utilizes the assets of the two Xeon Phi generations and we evaluate the merits of utilizing the KNL over its predecessor. In our case, an out-of-the-box transition on Knights Landing, offers on average 2.4x speed up while consuming 48% less energy than KNC.
KW - Computational neuroscience
KW - Intel xeon phi
KW - Knights landing
UR - http://www.scopus.com/inward/record.url?scp=85032681982&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-67630-2_27
DO - 10.1007/978-3-319-67630-2_27
M3 - Contribución a la conferencia
AN - SCOPUS:85032681982
SN - 9783319676296
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 363
EP - 375
BT - High Performance Computing - ISC High Performance 2017 International Workshops, DRBSD, ExaComm, HCPM, HPC-IODC, IWOPH, IXPUG, P^3MA, VHPC, Visualization at Scale, WOPSSS, Revised Selected Papers
A2 - Yokota, Rio
A2 - Kunkel, Julian M.
A2 - Taufer, Michela
A2 - Shalf, John
PB - Springer Verlag
Y2 - 18 June 2017 through 22 June 2017
ER -