Cosmological neutrino simulations at extreme scale  

作　　者：J.D.Emberson Hao-Ran Yu Derek Inman Tong-Jie Zhang Ue-Li Pen Joachim Harnois-Deraps Shuo Yuan Huan-Yu Teng Hong-Ming Zhu Xuelei Chen Zhi-Zhong Xing 

机构地区：[1]Canadian Institute for Theoretical Astrophysics,University of Toronto,M5S 3H8,ON,Canada [2]Department of Astronomy&Astrophysics,University of Toronto,Toronto,ON M5S 3H4,Canada [3]ALCF Division,Argonne National Laboratory,Lemont,IL 60439,USA [4]Kavli Institute for Astronomy&Astrophysics,Peking University,Beijing 100871,China [5]Department of Astronomy,Beijing Normal University,Beijing 100875,China [6]Department of Physics,University of Toronto,Toronto,ONtario M5S IA7,Canada [7]School of Physics and Electric Information,Shandong Provincial Key Laboratory of Biophysics,Dezhou University,Dezhou 253023,China [8]Dunlap Institute for Astronomy and Astrophysics,University of Toronto,Toronto,ON M5S 3H4,Canada [9]Canadian Institute for Advanced Research,Program in cosmology and Gravitation [10]Perimeter Institute for Theoretical Physics,Waterloo,ON,N2L 2Y5,Canada [11]Department of Physics&Astronomy,University of British Columbia,Vancouver,BC V6T 1Z1,Canada [12]Scottish University Physics Alliance,Institute for Astronomy,University of Edinburgh,EH93HJ,Scotland,UK [13]Department of Astronomy,Peking University,Beijing 100871,China [14]Key Laboratory for Computational Astrophysics,National Astronomical Observatories,Chinese Academy ofSciences,Beijing 100012,China [15]School of Physical Sciences,University of Chinese Academy of Sciences,Beijing 100049,China [16]Institute of High Energy Physics,Chinese Academy of Sciences,Beijing 100049,China

出　　处：《Research in Astronomy and Astrophysics》2017年第8期89-100,共12页天文和天体物理学研究（英文版）

基　　金：the Special Program for Applied Research on Super Computation of the NSFC-Guangdong Joint Fund(the second phase);supported under the U.S.Department of Energy contract DE-AC02-06CH11357;General Financial Grant No.2015M570884;Special Financial Grant No.2016T90009 from the China Postdoctoral Science Foundation;support from the European Commission under a Marie-Sklodwoska-Curie European Fellowship(EU project 656869);support from Mo ST 863 program 2012AA121701;NSFC grant 11373030;CAS grant QYZDJ-SSW-SLH017;supported by the National Natural Science Foundation of China(Grant Nos.11573006,11528306,10473002 and 11135009);the National Basic Research Program of China(973 program)under grant No.2012CB821804;the Fundamental Research Funds for the Central Universities;SciNet is funded by:the Canada Foundation for Innovation under the auspices of Compute Canada;the Government of Ontario;the Ontario Research Fund Research Excellence;the University of Toronto

摘　　要：Constraining neutrino mass remains an elusive challenge in modern physics.Precision measurements are expected from several upcoming cosmological probes of large-scale structure.Achieving this goal relies on an equal level of precision from theoretical predictions of neutrino clustering.Numerical simulations of the non-linear evolution of cold dark matter and neutrinos play a pivotal role in this process.We incorporate neutrinos into the cosmological N-body code CUBEP3M and discuss the challenges associated with pushing to the extreme scales demanded by the neutrino problem.We highlight code optimizations made to exploit modern high performance computing architectures and present a novel method of data compression that reduces the phase-space particle footprint from 24 bytes in single precision to roughly 9 bytes.We scale the neutrino problem to the Tianhe-2 supercomputer and provide details of our production run,named Tian Nu,which uses 86%of the machine(13 824 compute nodes).With a total of 2.97 trillion particles,Tian Nu is currently the world’s largest cosmological N-body simulation and improves upon previous neutrino simulations by two orders of magnitude in scale.We finish with a discussion of the unanticipated computational challenges that were encountered during the Tian Nu runtime.Constraining neutrino mass remains an elusive challenge in modern physics.Precision measurements are expected from several upcoming cosmological probes of large-scale structure.Achieving this goal relies on an equal level of precision from theoretical predictions of neutrino clustering.Numerical simulations of the non-linear evolution of cold dark matter and neutrinos play a pivotal role in this process.We incorporate neutrinos into the cosmological N-body code CUBEP3M and discuss the challenges associated with pushing to the extreme scales demanded by the neutrino problem.We highlight code optimizations made to exploit modern high performance computing architectures and present a novel method of data compression that reduces the phase-space particle footprint from 24 bytes in single precision to roughly 9 bytes.We scale the neutrino problem to the Tianhe-2 supercomputer and provide details of our production run,named Tian Nu,which uses 86%of the machine(13 824 compute nodes).With a total of 2.97 trillion particles,Tian Nu is currently the world’s largest cosmological N-body simulation and improves upon previous neutrino simulations by two orders of magnitude in scale.We finish with a discussion of the unanticipated computational challenges that were encountered during the Tian Nu runtime.

关 键 词：COSMOLOGY theory—large-scale structure of universe—methods NUMERICAL 

分 类 号：P141.8[天文地球—天体物理] P159[天文地球—天文学]