COSMOS - Framework for Combining Optimization and Simulation Software

被引:1
作者
Samoila, Cosmin-Gabriel [1 ]
Slusanschi, Emil-Ioan [1 ]
机构
[1] Univ Politehn Bucuresti, Dept Comp Sci & Engn, Bucharest, Romania
来源
2019 22ND INTERNATIONAL CONFERENCE ON CONTROL SYSTEMS AND COMPUTER SCIENCE (CSCS) | 2019年
关键词
high performance computing; optimization; simulation software; job scheduling;
D O I
10.1109/CSCS.2019.00035
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
A simulation software is a mathematical model translated into code and it describes a physical system. Given the fact that designing and writing simulation software requires a lot of knowledge about the system we want to model, we will focus on integrating optimization software with existing simulation programs for high energy particles and molecular dynamics. An optimization software is a special kind of software that targets applications and tunes their parameters and input accordingly to the desired result. By performing an optimization step, we are aiming either a lower run time or a better simulation precision. COSMOS is a framework that aims to provide three main functions: to optimize simulation output with respect to program input and parameters, to provide improved scheduling policies for all tasks of the simulation packages on HPC resources, and to offer enhanced sensitivity analysis of application parameters.
引用
收藏
页码:162 / 169
页数:8
相关论文
共 13 条
[1]  
Agrawal A., 2012, INT C MOD OPT COMP I
[2]  
Capdevielle J.N., 1998, CORSIKA MONTE CARLO
[3]   Versatile, scalable, and accurate simulation of distributed applications and platforms [J].
Casanova, Henri ;
Giersch, Arnaud ;
Legrand, Arnaud ;
Quinson, Martin ;
Suter, Frederic .
JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2014, 74 (10) :2899-2917
[4]  
Cherenkov P.A., 2007, VISIBLE EMISSION CLE, DOI [10.1070/PU2007v050n04ABEH006238/pdf, DOI 10.1070/PU2007V050N04ABEH006238/PDF]
[5]  
Fesefeldt H., 1980, PITHA8502
[6]  
Izakian H, 2009, NEURAL NETW WORLD, V19, P695
[7]  
Lang B., 2003, SCI PROG, V11
[8]  
Moore J.J., 1998, IEEE COMPUTATIONAL S, V5, P6875
[9]  
Pierog T., 2016, EXTENSIVE AIR SHOWER
[10]   EFCOSS: An Interactive Environment Facilitating Optimal Experimental Design [J].
Rasch, Arno ;
Buecker, H. Martin .
ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 2010, 37 (02)