Proactive resource provisioning

Authors:
Eric Chi;Michael Fu;Jean Walrand
Affiliations:
Department of Electrical Engineering and Computer Sciences, University of California at Berkeley, 211 Cory Hall #1772, Berkeley, CA 94720-1772, USA;Department of Electrical Engineering and Computer Sciences, University of California at Berkeley, 211 Cory Hall #1772, Berkeley, CA 94720-1772, USA;Department of Electrical Engineering and Computer Sciences, University of California at Berkeley, 211 Cory Hall #1772, Berkeley, CA 94720-1772, USA
Venue:
Computer Communications
Year:
2004

Citing 6
Cited 1

AQUILA: adaptive resource control for QoS using an IP-based layered architecture

IEEE Communications Magazine
Cadenus: creation and deployment of end-user services in premium IP networks

IEEE Communications Magazine
Dynamic staffing in a telephone call center aiming to immediately answer all calls

Operations Research Letters
Distributed admission control

IEEE Journal on Selected Areas in Communications
Internet QoS: a big picture

IEEE Network: The Magazine of Global Internetworking
A policy-based quality of service management system for IP DiffServ networks

IEEE Network: The Magazine of Global Internetworking

Anatomy of delay performance for the strict priority scheduling scheme in multi-service Internet

Computer Communications

Quantified Score

Hi-index	0.24

Visualization

Abstract

Allocating resources in networks to QoS flows may require undesirable delays or costs. We consider a dynamic Service Level Agreement (SLA) negotiation scheme between peer autonomous systems (ASes) that implement DiffServ per domain behaviors. For concreteness, we tailor our scheme to account for the needs of Voice over IP transport across multiple ASes. We present a heuristic but computationally simple and distributed scheme that uses traffic statistics to forecast the near-future demand. Our scheme provides a statistical guarantee on the renegotiation frequency, to manage adjustment costs, and blocking probability, to manage penalties for undesirable adjustment delays. Simulations demonstrate that this scheme achieves more efficient usage of the reserved bandwidth than would be accomplished by using the de facto static SLA negotiation schemes.