Abstract
There is an increasing demand for Internet core nodes to have (1) quality-of-service (QoS) guarantees and (2) higher switching capacity. The combined input and output queued (CIOQ) switching has been shown to be a promising solution to meet such demand. However, many of the proposed scheduling algorithms for a CIOQ switch cannot be practically implemented due to problems from scalability and complexity. This paper shows how the virtual time reference system (VTRS) can be applied to packet scheduling in a CIOQ switch. The VTRS is a unifying scheduling framework to provide scalable support for guaranteed services. In the context of packet scheduling for a CIOQ switch, we show that the use of VTRS can eliminates both the scalability and complexity problems associated with many of the scheduling algorithms proposed in the literature. More important, we show that in term of of providing end-to-end guaranteed service, packet scheduling using VTRS for a CIOQ switch has the same performance as an output queued (OQ) switch employing weighted fair queueing (WFQ) scheduler.
Original language | English (US) |
---|---|
Title of host publication | Proceedings - 7th International Conference on Parallel and Distributed Systems |
Subtitle of host publication | Workshops |
Editors | Yoshitaka Shibata |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
Pages | 239-248 |
Number of pages | 10 |
ISBN (Electronic) | 0769505716, 9780769505718 |
DOIs | |
State | Published - 2000 |
Event | 7th International Conference on Parallel and Distributed Systems, ICPADS 2000 - Iwate, Japan Duration: Jul 4 2000 → Jul 7 2000 |
Publication series
Name | Proceedings - 7th International Conference on Parallel and Distributed Systems: Workshops |
---|
Other
Other | 7th International Conference on Parallel and Distributed Systems, ICPADS 2000 |
---|---|
Country/Territory | Japan |
City | Iwate |
Period | 7/4/00 → 7/7/00 |
Bibliographical note
Publisher Copyright:© 2000 IEEE.
Keywords
- Guaranteed service
- Packet switching
- QoS
- Scalability
- Scheduling
- Weighted fair queueing