ABSTRACT
Interactive multimedia sessions cause high variations in workload due to presenting different media streams at different points in time. As users interact, the workload variations can not be predicted precisely. But admission control mechanisms need to at least estimate the workload in order to provide Quality of Service. In this paper, we investigate two approaches to estimate workloads and on this basis introduce admission control mechanisms that give stochastic QoS guarantees for such sessions. The observation based approach uses bookkept workload data for the prediction. The model based approach employs knowledge about usage patterns and media streams. For both approaches we introduce a uniform stochastic admission control criterion. Furthermore, we illustrate the relative benefits of these approaches for various session scenarios by means of simulations.
- P. Agrawal, J.-C. Chen, and C. J. Sreenan. Use of statistical methods to reduce delays for media playback buffering. In ICMCS, 1998.]] Google ScholarDigital Library
- N. H. Balkir and G. Ozsoyoglu. Delivering presentations from multimedia servers. VLDB Journal. Special Issue on Multimedia Databases, 1998.]] Google ScholarDigital Library
- M. Friedrich, S. Hollfelder, and K. Aberer. Stochastic resource prediction and admission for interactive sessions on multimedia servers. In ACM Multimedia, 2000.]] Google ScholarDigital Library
- S. Hollfelder and K. Aberer. An admission control framework for applications with variable consumption rates in client-pull architectures. In MIS, 1998.]] Google ScholarDigital Library
- L. Kleinrock. Queueing Systems, Volume I: Theory. Wiley, 1975.]] Google ScholarDigital Library
- J. Qiu and E. W. Knightly. Measurement-based admission control with aggregate traffic envelopes. IEEE/ACM Transactions on Networking, 9(2), 2001.]] Google ScholarDigital Library
- R. Rangaswami, Z. Dimitrijevic, E. Chang, and S. G. Chan. Fine-grained device management in an interactive media server. to appear in IEEE Transactions on Multimedia, 2003.]]Google ScholarDigital Library
- H. C. Tijms. Stochastic Models. An Algorithmic Approach. Wiley series in probability and mathematical statistics. Wiley, 1994.]]Google Scholar
- H. M. Vin, P. Goyal, A. Goyal, and A. Goyal. A statistical admission control algorithm for multimedia servers. In ACM Multimedia, 1994.]] Google ScholarDigital Library
- M. Y.Y.Leung, J. C. Lui, and L. Golubchik. Buffer and I/O resource pre-allocation for implementing batching and buffering techniques for video-on-demand systems. In ICDE, 1997.]] Google ScholarDigital Library
- W. Zhao and S. K. Tripathi. A resource reservation scheme for synchronized distributed multimedia sessions. Multimedia Tools and Applications, 7(1/2), 1998.]] Google ScholarDigital Library
Index Terms
- Observation based vs. model based admission control for interactive multimedia sessions
Recommendations
Stochastic resource prediction and admission for interactive sessions on multimedia servers
MULTIMEDIA '00: Proceedings of the eighth ACM international conference on MultimediaIn highly interactive multimedia applications startup latency is significant, and may negatively impact performance and Quality of Service (QoS). To avoid this, our approach is to admit whole multimedia sessions instead of single media streams. For the ...
Measurement-based admission control at edge routers
It is very important to allocate and manage resources for multimedia traffic flows with real-time performance requirements in order to guarantee quality of service (QoS). In this paper, we develop a scalable architecture and an algorithm for admission ...
Measurement-Based Admission Control: A Large Deviations Approach for Bufferless Multiplexers
ISCC '00: Proceedings of the Fifth IEEE Symposium on Computers and Communications (ISCC 2000)In order to provide Quality of Service (QoS) assurances networks perform call admission control before accepting a new connection. Rather than relying on \apriori traffic descriptors (such as leaky buckets), which often poorly characterize the actual ...
Comments