Abstract:High speed and multimedia are two trends of the future network. While today's computer networks support only best-effort service, future packet switching integrated service networks will have to support real-time communication services that allow clients to transport information with performance guarantees expressed in terms of delay, delay jitter, throughput, and loss rate. To provide reliable end-to-end quality of service guarantees to the network users, it is necessary to introduce per-flow queuing scheduling algorithms to networks nodes. The Generalized Processor Sharing (GPS) discipline is proven to have desirable properties which is the basis for supporting both the guaranteed service traffic and the best-effort service traffic. Among all the packet approximation algorithms of GPS, WFQ, also known as Packet Generalized Processor Sharing (PGPS), has been considered to be the best one in terms of accuracy. WF2Q+ is an improved algorithm of WFQ, which has the most excellent performance and can be easily implemented at the same time. But the complexity of WF2Q+ implementing structure and sorting operations make it very difficult to be implemented in a high-speed environment. By observing the traffic on the Internet, we can easily find the statistical characteristics of the Internet traffic. In this paper, we provide a high-speed hardware implementation of WF2Q+ algorithm based on statistical shift registering sorting architectures. By making full use of the statistic information of each queue, high-speed and complete sorting operations in sense of statistics can be implemented with less hardware resource. We also implemented our designs on the Field Programmable Gate Array (FPGA) chips by using VHDL hardware description language. According to the implementation results of FPGA, this architecture is fully capable of OC-48 (2.4Gbps) operation in high-speed IP routers. Our simulation results also proved that the scheduling effect of the algorithm we implemented is almost the same as the standard WF2Q+ algorithm.

A Novel Processor Scheduling Schema in Programmable Router Based on Queue Length Thresholds

Processor Scheduling in Programmable Router Based on Queue Length Thresholds

A Period-Aware Routing Method for IEEE 802.1qbv TSN Networks

Enhancing Router QoS Through Job Scheduling with Weighted Shortest Processing Time-Adjusted.

Implementation of Packet-Fair Queuing Algorithms in High Speed Routers Based on Discrete Rates and Packet Lengths Group Architectures

Design of Packet-Fair Queuing Schedulers in Router

An Efficient Scheduling Mechanism With Flow-Based Packet Reordering In A High-Speed Network Processor

Study on a Packet Scheduling Algorithm in High Performance Routers

Multiple-threshold Based Scheduling Algorithm for High Performance Input Queued Switches

Neural Network Based Scheduling for Variable-Length Packets in Gigabit Router with Crossbar Switch Fabric and Input Queuing

Traffic-Based Scheduling Mechanism And Performance Analysis In Ip/Atm Router

Variable-length Packet Scheduling Algorithm PDMCRR

Minimizing the Communication Overhead of Iterative Scheduling Algorithms for Input-Queued Switches.

A Scalable Per-flow Priority Scheduling Scheme for High-Speed Network

A Novel Routing Algorithm for the Acceleration of Flow Scheduling in Time-Sensitive Networks

Load Balanced Packet Scheduling Algorithm of Terabit Router

Implementing Fair Queuing Algorithms in High Speed Routers Based on

Research of Router Architecture Based on Network Processor

Dynamic Queuing Sharing Mechanism for Per-Flow Quality of Service Control

Vpifo: Virtualized Packet Scheduler for Programmable Hierarchical Scheduling in High-Speed Networks

Data Path Processing in Fast Programmable Routers