Optimal Run Length for Discrete-event Distributed Cluster-based Simulations

In scientific simulations the results generated usually come from a stochastic process. New solutions with the aim of improving these simulations have been proposed, but the problem is how to compare these solutions since the results are not deterministic. Consequently how to guarantee that the outp...

Descripción completa

Detalles Bibliográficos
Autores: Borges, Francisco|||0000-0003-4951-0522, Gutiérrez Millà, Albert|||0000-0002-6242-9414, Suppi, Remo|||0000-0002-0373-8292, Luque, Emilio|||0000-0002-2884-3232
Tipo de recurso: artículo
Fecha de publicación:2014
País:España
Institución:Universitat Autònoma de Barcelona
Repositorio:Dipòsit Digital de Documents de la UAB
Idioma:inglés
OAI Identifier:oai:ddd.uab.cat:306173
Acceso en línea:https://ddd.uab.cat/record/306173
https://dx.doi.org/urn:doi:10.1016/j.procs.2014.05.007
Access Level:acceso abierto
Palabra clave:Parallel and distributed simulation
Parallel discrete-event simulation
High performance distributed simulation
Output analysis
Run length
Transient state
Steady state
Descripción
Sumario:In scientific simulations the results generated usually come from a stochastic process. New solutions with the aim of improving these simulations have been proposed, but the problem is how to compare these solutions since the results are not deterministic. Consequently how to guarantee that the output results are statistically trusted. In this work we apply a statistical approach in order to define the transient and steady state in discrete event distributed simulation. We used linear regression and batch method to find the optimal simulation size. As contributions of our work we can enumerate: we have applied and adapted the simple statistical approach in order to define the optimal simulation length; we propose the approximate approach to normal distribution instead of generate replications sufficiently large; and the method can be used in other kind of non-terminating science simulations where the data either have a normal distribution or can be approximated by a normal distribution.