3rd International IEEE/Create-Net Workshop on Networks for Grid Applications

Research Article

A Flexible Advance Reservation Model for Multi-Domain WDM Optical Networks

  • @INPROCEEDINGS{10.4108/gridnets.2006.13,
        author={Eric He and Xi Wang and Jason Leigh},
        title={A Flexible Advance Reservation Model for Multi-Domain WDM Optical Networks},
        proceedings={3rd International IEEE/Create-Net Workshop on Networks for Grid Applications},
        publisher={IEEE},
        proceedings_a={GRIDNETS},
        year={2006},
        month={10},
        keywords={Advance Reservation  Interdomain  Intradomain  LambdaGrid  Lightpath  RWA},
        doi={10.4108/gridnets.2006.13}
    }
    
  • Eric He
    Xi Wang
    Jason Leigh
    Year: 2006
    A Flexible Advance Reservation Model for Multi-Domain WDM Optical Networks
    GRIDNETS
    IEEE
    DOI: 10.4108/gridnets.2006.13
Eric He1,*, Xi Wang1, Jason Leigh1
  • 1: Electronic Visualization Laboratory, Department of Computer Science, University of Illinois at Chicago
*Contact email: eric@evl.uic.edu

Abstract

Advance reservation is a mechanism to guarantee the availability of resources when they are needed. In the context of LambdaGrid, this mechanism is used to provide data-intensive applications with the needed deterministic network quality of service to transport data between grid instruments, high-performance storage systems, compute clusters and visualization systems. Flexible scheduling affords users greater convenience while also improving resource utilization and acceptance rate. In this paper we propose a flexible advance reservation model (FARM) and describe how to implement this model in the meta-scheduling problem. Then we extend this methodology to the cross-domain lightpath reservation problem by incorporating routing and wavelength assignment algorithms. Next, we present the architecture, implementation and APIs of a coordinated interdomain and intradomain optical control plane called AR-PIN/PDC, which is capable of flexible advance reservations. Our simulation results show that by relaxing the reservation time constraint, the acceptance rate and resource utilization can be improved dramatically. Through simulations, we also analyze the impact of advance reservations on immediate reservations and conclude that both AR and IR requests need admission control algorithms in order to let both types of reservations coexist and use resource properly.