On the Convergence of Multiclass Queueing Networks in Heavy Traffic (Classic Reprint) - Couverture souple

Dai, J. G.

 
9781334017919: On the Convergence of Multiclass Queueing Networks in Heavy Traffic (Classic Reprint)

Synopsis

Rethinking how we model busy networks under heavy load and what that means for real‑world performance.

This book examines multiclass open queueing networks and shows that a long‑held Brownian approximation does not always apply. It offers a careful look at when standard limits fail and why more general models may be needed.

In clear terms, the work frames the problem of predicting system behavior as demand grows without bound. It introduces a pseudo heavy traffic limit and demonstrates that not all networks converge to a continuous limit. Readers will see how specific network structures can defy expected Brownian models, and why this matters for analysis and design.
  • Introduces the heavy traffic framework and key terms like workload and idleness in multiclass networks.
  • Shows that certain two‑ and three‑station networks do not converge to the standard Brownian model.
  • Explains the idea of a pseudo heavy traffic limit and its implications for modeling accuracy.
  • Discusses how these results affect the use of existing approximation methods and routing considerations.
Ideal for readers of advanced queueing theory, stochastic networks, and researchers seeking a rigorous assessment of when Brownian approximations are reliable.

Les informations fournies dans la section « Synopsis » peuvent faire référence à une autre édition de ce titre.

Autres éditions populaires du même titre