Dissertation > Excellent graduate degree dissertation topics show
Research on Switching Architectures Based on Multi-hop Routing
Author: ZhengDeRen
Tutor: WangBinQiang
School: PLA Information Engineering University
Course: Communication and Information System
Keywords: Multi-port output Switching Architecture Scheduling Round Robin
CLC: TN915.02
Type: Master's thesis
Year: 2009
Downloads: 10
Quote: 0
Read: Download Dissertation
Abstract
|
Conventional routing and switching technology is designed mainly based on the single forwarding path, giving only one output port to the packets entered the switching device. The contention of packets with the same output port is congestion-proneness, and deteriorating the performance of switching device or even the whole network. So traditional switching scheme has greatly restricted the performance of network.Multi-hop routing mechanism has achieved great development recently, which leads the cells have multiple output ports. So with the new switching scheme study, the scheduling congestion of the same output port can be highly reduced. Based on the researching result of multi-hop routing, the dissertation presents a queues’length dynamic balancing algorithm and a new switching fabric. Combined with the project“A Dynamic Balancing Routing and Switching Mechanism Based on Multi-Hop”supported by the National Program 863, it also gives a project implement scheme supporting multiple output ports.Its main work is outlined as follows:? With the idea of routing and switching cooperating design, the dissertation proposes a multi-port output switching mechanism, then comes to a conclusion that the switching performance improves greatly when the number of arriving cell’s output port is been set to 3, finally validates the conclusion with the Expand Switching Performance Evaluation System.? Combined with the characteristic of the Input queued switch, the dissertation presents a queues’length dynamic balancing matching algorithm. The algorithm balances the queues both by input scheme and by scheduling policy. The simulation results show that the throughput can be increased more than 25% compared to iSLIP under non-uniform traffics with one iteration, and the performance of delay is also improved greatly.? The dissertation builds a switch with parallel memory scheme(PMS). With the flow model techniques we prove that the PMS can achieve a throughput of 100% without speedup. Based on the PMS, we present a parallel distributed compensating round robin algorithm supporting multiple output ports. The algorithm can schedule the queues of different output port independently with parallel flow manner. The simulation results indicate that the PMS switch with PDCRR algorithm can obtain better performance.? According to the requirements of the project and considering systematic performance and implement difficulty, a system solution has been provided based on the PMS and PDCRR. Simulation of FPGA shows that the scheme completely meets the project requirements.
|
Related Dissertations
- Research of Scheduling Algorithm Based on Hybrid Adaptive Genetic Algorithm in Computing Grid,TP393.09
- Public Transport Optimal Dispatching Based on the Genetic-Newton Algorithm,TP18
- On Data Scheduling Strategies of Serving Peers in P2P VoD Systems,TN948.64
- Research and Implementation of job scheduling algorithm in the Linux cluster environment,TP301.6
- Three Gorges Cascade Reservoir Scheduling with Fuzzy Optimization Method,TV697.1
- Cooperative Optimization Scheduling with Application to Multi-Reservoir System During Non-Flood Period,TV697.11
- Design and Implementation of Market Benchmark Reference System,TP311.52
- MTO supply chain 3PL Transportation Scheduling Problem,F224
- Based on distributed real-time database transaction scheduling policy Research and Improvement,TP311.13
- Fault-tolerant real-time systems based on energy-efficient scheduling algorithm,TP316.2
- Particle Swarm Optimization in Reservoir Flood Dispatching Research,TP301.6
- Embedded real-time operating system ARTs-OS 's EDF scheduling improvements,TP316.2
- Overload virtual machine performance improvement under MPI communication method,TP302
- Lock-based multiprocessor -aware scheduling system VCPU,TP332
- SOC platform for advanced control design and implementation,TP273
- Temperature limits based on multi-core systems saving scheduling algorithm,TP332
- Broadband wireless communications multi-user scheduling and diversity techniques,TN92
- Multi-cell interference on the downlink MIMO communication system modeling and collaborative scheduling,TN929.5
- P2P streaming media technology based IPTV System,TP393.02
- Research on Shipping Route Scheduling for Container Liner Under Emergency Situation,F224
- Models and Opitimization Algorithms for Production Schedling Problems Based on Energy Saving,F206
CLC: > Industrial Technology > Radio electronics, telecommunications technology > Communicate > Communication network > General issues > Communication network structure and design
© 2012 www.DissertationTopic.Net Mobile
|