Adaptive multi-GPU Exchange Monte Carlo for the 3D Random Field Ising Model

doi:http://dx.doi.org/10.1016/j.cpc.2016.04.007

CORC > 理论物理研究所 > 中国科学院理论物理研究所 > 理论物理所科研产出 > 理论物理所1978-2010年知识产出

	Adaptive multi-GPU Exchange Monte Carlo for the 3D Random Field Ising Model
	Navarro, CA ; Huang, W 2; Deng, YJ
刊名	COMPUTER PHYSICS COMMUNICATIONS
	2016
卷号	205 页码:48-60
关键词	GPU computing STATE POTTS-MODEL Adaptive temperatures CLUSTER ALGORITHM Exchange Monte Carlo algorithm CRITICAL-BEHAVIOR Random Field Ising Model SPIN-GLASSES SIMULATION
DOI	http://dx.doi.org/10.1016/j.cpc.2016.04.007
英文摘要	This work presents an adaptive multi-GPU Exchange Monte Carlo approach for the simulation of the 3D Random Field Ising Model (RFIM). The design is based on a two-level parallelization. The first level, spin-level parallelism, maps the parallel computation as optimal 3D thread-blocks that simulate blocks of spins in shared memory with minimal halo surface, assuming a constant block volume. The second level, replica-level parallelism, uses multi-GPU computation to handle the simulation of an ensemble of replicas. CUDA's concurrent kernel execution feature is used in order to fill the occupancy of each GPU with many replicas, providing a performance boost that is more notorious at the smallest values of L. In addition to the two-level parallel design, the work proposes an adaptive multi-GPU approach that dynamically builds a proper temperature set free of exchange bottlenecks. The strategy is based on mid-point insertions at the temperature gaps where the exchange rate is most compromised. The extra work generated by the insertions is balanced across the GPUs independently of where the mid-point insertions were performed. Performance results show that spin-level performance is approximately two orders of magnitude faster than a single-core CPU version and one order of magnitude faster than a parallel multi-core CPU version running on 16-cores. Multi-GPU performance is highly convenient under a weak scaling setting, reaching up to 99% efficiency as long as the number of GPUs and L increase together. The combination of the adaptive approach with the parallel multi-GPU design has extended our possibilities of simulation to sizes of L = 32, 64 for a workstation with two GPUs. Sizes beyond L = 64 can eventually be studied using larger multi-GPU systems. (C) 2016 Elsevier B.V. All rights reserved.
学科主题	Computer Science ; Physics
语种	英语
资助机构	Nvidia GPU Research Center at the Department of Computer Science (DCC) of University of Chile ; Supercomputing Center of University of Science and Technology of China ; FONDECYT [3160182] ; CONICYT, Chile ; National Science Foundation of China (NSFC) [11275185] ; Open Project Program of State Key Laboratory of Theoretical Physics, Institute of Theoretical Physics, Chinese Academy of Sciences, China [Y5KF191CJ1]
内容类型	期刊论文
源URL	[http://ir.itp.ac.cn/handle/311006/23030]
专题	理论物理研究所_理论物理所1978-2010年知识产出
作者单位	1.Univ Sci & Technol China, Dept Modern Phys, Hefei Natl Lab Phys Sci Microscale, Hefei 230027, Peoples R China 2.[Navarro, Cristobal A.] Univ Austral Chile, Inst Informat, Valdivia, Chile 3.Chinese Acad Sci, State Key Lab Theoret Phys, Inst Theoret Phys, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Navarro, CA,Huang, W,Deng, YJ. Adaptive multi-GPU Exchange Monte Carlo for the 3D Random Field Ising Model[J]. COMPUTER PHYSICS COMMUNICATIONS,2016,205:48-60.
APA	Navarro, CA,Huang, W,&Deng, YJ.(2016).Adaptive multi-GPU Exchange Monte Carlo for the 3D Random Field Ising Model.COMPUTER PHYSICS COMMUNICATIONS,205,48-60.
MLA	Navarro, CA,et al."Adaptive multi-GPU Exchange Monte Carlo for the 3D Random Field Ising Model".COMPUTER PHYSICS COMMUNICATIONS 205(2016):48-60.