Backpressure control is one of several effective traffic control schemes for ATM-LAN. Comparing an input queuing switch with an output queuing switch, backpressure control in a congested input queuing switch has the ability to selectively control the output port of the front stage of the input queui
Performance evaluation of backpressure congestion control in the multistage model
β Scribed by Kenji Kawahara; Takashi Watanabe; Yuji Oie
- Publisher
- John Wiley and Sons
- Year
- 2000
- Tongue
- English
- Weight
- 203 KB
- Volume
- 83
- Category
- Article
- ISSN
- 8756-6621
No coin nor oath required. For personal study only.
β¦ Synopsis
The combination of reactive control with CAC or UPC has been proposed as the congestion control method in ATM networks. As one of these methods, the authors researched the backpressure congestion control method and analyzed the basic performance of the two-stage model. The model consists of the congested node and its next node; it was shown that the effect of congestion can be reduced in traffic that does not pass through the congested node by adding pushout rules to the next node.
In this article, the discussion is extended to a multistage model that includes source, relay, and congested nodes, and the propagation time of congestion in the network, the packet loss probability, and the average packet transmission delay time are estimated by computer simulations. It is shown that congestion can be reduced and the propagation of the congestion state also can be decreased by the addition of pushout rules to backpressure congestion control.
π SIMILAR VOLUMES
The characterization of a stream's low-Β―ow regime is required for ecological purposes, water quality studies and various other water projects. If observed stream Β―ow records are insucient, low-Β―ow characteristics may need to be estimated from simulated daily stream Β―ow time-series. The model employe
## Abstract Recent technical developments in computer hardware and software have meant that humanβmachine systems can be automated in many respects. If automation fails, however, human operators can have difficulty in recognizing the existence of a problem, identifying what has failed, and taking c
## Abstract The use of multiple models for adaptively controlling an unknown continuousβtime linear system was proposed in Narendra and Balakrishnan (__IEEE Transactions on Automatic Control__ 1994; **39**(9):1861β1866). and discussed in detail in Narendra and Xiang (__IEEE Transactions on Automati
This project compared the effect of formulation variables on the dissolution performance of model Biopharmaceutics Classification System (BCS) Class II drugs from hard gelatin capsules using statistical analysis and Bayesian networks. The drugs chosen for this study were carbamazepine (CAR), chlorpr