Unleashing Simple Pendulum Dynamics with Posit Arithmetic

被引:0
作者
Aldhapati, Avinash [1 ]
Kumar, Ashwini Jaya [1 ]
Subramanian, Rajaraman [1 ]
机构
[1] Calligo Technol Pvt Ltd, Bengaluru, India
来源
NEXT GENERATION ARITHMETIC, CONGA 2024 | 2024年 / 14666卷
关键词
Posits; IEEE; 754; Simple Pendulum; Simulations; high-performance computing; Error; Precision; SIMULATION;
D O I
10.1007/978-3-031-72709-2_5
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Simulations provide a powerful means to explore, analyze, and understand complex systems, allowing us to make informed decisions across various domains. Fields like genomics, physics, and climate science and many more heavily rely on simulations to produce valuable information but the Memory Wall, the gap between computation speed and data access, poses challenges, especially these days when massive amount of data is produced and quick access to it is of high priority. While solutions exist, they often require infrastructure changes or apply to specific algorithms only. Posit, a novel datatype, matches standard sizes but with higher precision. Analyzing systems like a pendulum with posits reveals lower errors compared to floats, enhancing accuracy within memory constraints. Posits excel in capturing system dynamics, surpassing competitors of similar sizes. They enable better predictions of complex simulations, advancing our understanding of natural phenomena while addressing memory limitations.
引用
收藏
页码:104 / 124
页数:21
相关论文
共 29 条
[1]   The impact of finite precision arithmetic and sensitivity on the numerical solution of partial differential equations [J].
Allen, E ;
Burns, J ;
Gilliam, D ;
Hill, J ;
Shubov, V .
MATHEMATICAL AND COMPUTER MODELLING, 2002, 35 (11-12) :1165-1195
[2]   A review of parallel computing applications in calibrating watershed hydrologic models [J].
Asgari, Marjan ;
Yang, Wanhong ;
Lindsay, John ;
Tolson, Bryan ;
Dehnavi, Maryam Mehri .
ENVIRONMENTAL MODELLING & SOFTWARE, 2022, 151
[3]   High-precision computation: Mathematical physics and dynamics [J].
Bailey, D. H. ;
Barrio, R. ;
Borwein, J. M. .
APPLIED MATHEMATICS AND COMPUTATION, 2012, 218 (20) :10106-10121
[4]  
Baker G.L., 2005, The Pendulum: A Case Study in Physics
[5]   Evaluation of climate models using palaeoclimatic data [J].
Braconnot, Pascale ;
Harrison, Sandy P. ;
Kageyama, Masa ;
Bartlein, Patrick J. ;
Masson-Delmotte, Valerie ;
Abe-Ouchi, Ayako ;
Otto-Bliesner, Bette ;
Zhao, Yan .
NATURE CLIMATE CHANGE, 2012, 2 (06) :417-424
[6]   Simulation-based inference in particle physi [J].
Brehmer, Johann .
NATURE REVIEWS PHYSICS, 2021, 3 (05) :305-305
[7]   Evaluating the Numerical Stability of Posit Arithmetic [J].
Buoncristiani, Nicholas ;
Shah, Sanjana ;
Donofrio, David ;
Shalf, John .
2020 IEEE 34TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM IPDPS 2020, 2020, :612-621
[8]  
callibri, About us
[9]   Multi-agent power market simulation using EMCAS [J].
Conzelmann, G ;
Boyd, G ;
Koritarov, V ;
Veselka, T .
2005 IEEE POWER ENGINEERING SOCIETY GENERAL MEETING, VOLS, 1-3, 2005, :2829-2834
[10]   Posits: the good, the bad and the ugly [J].
de Dinechin, Florent ;
Forget, Luc ;
Muller, Jean-Michel ;
Uguen, Yohann .
CONFERENCE FOR NEXT GENERATION ARITHMETIC 2019 (CONGA), 2019,