Advance reservations for predictive service in the Internet
Multimedia Systems - Special issue on the fifth workshop on network and operating system support for digital audio and video 1995 (NOSSDAV)
Competitive dynamic bandwidth allocation
PODC '98 Proceedings of the seventeenth annual ACM symposium on Principles of distributed computing
Resource sharing for book-ahead and instantaneous-request calls
IEEE/ACM Transactions on Networking (TON)
Issues ofReserving Resources in Advance
NOSSDAV '95 Proceedings of the 5th International Workshop on Network and Operating System Support for Digital Audio and Video
Design and evaluation of an advance reservation protocol on top of RSVP
BC '98 Proceedings of the IFIP TC6/WG6.2 Fourth International Conference on Broadband Communications: The future of telecommunications
LCN '00 Proceedings of the 25th Annual IEEE Conference on Local Computer Networks
Scheduling with Advanced Reservations
IPDPS '00 Proceedings of the 14th International Symposium on Parallel and Distributed Processing
ICDCS '01 Proceedings of the The 21st International Conference on Distributed Computing Systems
Hi-index | 0.00 |
In order to provide the QoS (Quality of Service) effectively to the network-based media applications consuming large bandwidth (e.g., video conferencing), the service providers of QoS-provisioned networks should allow customers to reserve their resources in advance. The service provider, however, should also provide conventional reservations (generally known as immediate reservations) that do not specify session duration. To adaptively control the amount of sharable resources between resource partitions for the respective immediate and advance reservations, in this paper, we propose a balanced revenue-based resource sharing scheme that maintains expected revenue within a pre-defined range while minimizing the management overhead. By analyzing the relationship between the revenue and the amount of resources under conflict (i.e., immediately reserved flows should be preempted to make room for advance reservation flows), we control the trade-off. That is, a resource boundary between two types of reservations is dynamically adjusted by a weighting parameter that represents the sharing modes for the expected revenue: aggressive and conservative. Network simulation results show that the proposed scheme exhibits enhanced performance (i.e., stabilized revenue and low management overhead) in comparison to alternative schemes, especially when the demand for two types of reservations varies.