Ayuda
Ir al contenido

Dialnet


Performance modeling of a pro-active multi-tier dynamic scheduling algorithm with threhold deriviations

  • Autores: Subramaniam K. Shamala, Mohd Yazid Saman, Mohamed Othman, Rozita Johari
  • Localización: International journal of the computer, the internet and management, ISSN 0858-7027, Vol. 9, Nº. 3, 2001, págs. 21-35
  • Idioma: inglés
  • Texto completo no disponible (Saber más ...)
  • Resumen
    • The Internet has revolutionized the culture of computing and telecommunications. The emergence of newly designed real-time multimedia applications is creating a pressing need to redesign and revamp the Internet architectures, algorithms and protocols.

      Scheduling algorithms are an integral component in determining the performance of packet-switched networks. Extensive research have been performed to derive the ideal correlation between scheduling algorithms, resource management and admission control. In this study, two new pro-active dynamic multi-tier scheduling algorithms for output-buffered switches with a priority and threshold derivation mechanisms are proposed. The two models are designed on the basis of the predictive service model. The proposed bandwidth and buffer management schemes are performed dynamically. A pro-active admission control algorithm is also proposed. The admission control algorithm utilizes a recursive formula to compute the probability of a packet incurring an estimated delay exceeding the stipulated service requirements. The second model integrates an additional threshold feature in the dynamic and pro-active compositions. The performance of the two proposed multi-tier scheduling algorithm is compared with the OCcuPancy_Adjusting (OCP_A) scheduling algorithm via extensive discrete-event simulation models. The simulation results demonstrate that the proposed multi-tier dynamic scheduling algorithms provide significant improvements as compared to the OCP_A.


Fundación Dialnet

Dialnet Plus

  • Más información sobre Dialnet Plus

Opciones de compartir

Opciones de entorno