Convolution algorithm for normalization constant evaluation in queuing system with random requirements

We suggest a convolution algorithm for calculating the normalization constant for stationary probabilities of a multiserver queuing system with random resource requirements. Our algorithm significantly reduces computing time of the stationary probabilities and system characteristics such as blocking probabilities and average number of occupied resources. The algorithm aims to avoid calculation of k-fold convolutions and reasonably use memory resources. © 2017 Author(s).

Авторы
Samouylov K. 1 , Sopin E. 1, 2 , Vikhrova O. 1 , Shorgin S.2
Сборник материалов конференции
Язык
Английский
Статус
Опубликовано
Номер
090004
Том
1863
Год
2017
Организации
  • 1 Applied Probability and Informatics Department, RUDN University, Moscow, Russian Federation
  • 2 Institute of Informatics Problems, Federal Research Center Computer Science and Control, Moscow, Russian Federation
Дата создания
19.10.2018
Дата изменения
19.10.2018
Постоянная ссылка
https://repository.rudn.ru/ru/records/article/record/5410/