Articles via Databases
Articles via Journals
Online Catalog
E-books
Research & Information Literacy
Interlibrary loan
Theses & Dissertations
Collections
Policies
Services
About / Contact Us
Administration
Littman Architecture Library
This site will be removed in January 2019, please change your bookmarks.
This page will redirect to https://digitalcommons.njit.edu/dissertations/560 in 5 seconds

The New Jersey Institute of Technology's
Electronic Theses & Dissertations Project

Title: Joint buffer management and scheduling for input queued switches
Author: Liu, Dequan
View Online: njit-etd2003-024
(xi, 71 pages ~ 3.5 MB pdf)
Department: Department of Electrical and Computer Engineering
Degree: Doctor of Philosophy
Program: Electrical Engineering
Document Type: Dissertation
Advisory Committee: Hou, Edwin (Committee co-chair)
Ansari, Nirwan (Committee co-chair)
Tekinay, Sirin (Committee member)
Papavassiliou, Symeon (Committee member)
Liu, Jiming (Committee member)
Date: 2003-01
Keywords: Input queued switches
Buffer management
Availability: Unrestricted
Abstract:

Input queued (IQ) switches are highly scalable and they have been the focus of many studies from academia and industry. Many scheduling algorithms have been proposed for IQ switches. However, they do not consider the buffer space requirement inside an IQ switch that may render the scheduling algorithms inefficient in practical applications.

In this dissertation, the Queue Length Proportional (QLP) algorithm is proposed for IQ switches. QLP considers both the buffer management and the scheduling mechanism to obtain the optimal allocation region for both bandwidth and buffer space according to real traffic load. In addition, this dissertation introduces the Queue Proportional Fairness (QPF) criterion, which employs the cell loss ratio as the fairness metric. The research in this dissertation will show that the utilization of network resources will be improved significantly with QPF. Furthermore, to support diverse Quality of Service (QoS) requirements of heterogeneous and bursty traffic, the Weighted Minmax algorithm (WMinmax) is proposed to efficiently and dynamically allocate network resources.

Lastly, to support traffic with multiple priorities and also to handle the decouple problem in practice, this dissertation introduces the multiple dimension scheduling algorithm which aims to find the optimal scheduling region in the multiple Euclidean space.


If you have any questions please contact the ETD Team, libetd@njit.edu.

 
ETD Information
Digital Commons @ NJIT
Theses and DIssertations
ETD Policies & Procedures
ETD FAQ's
ETD home

Request a Scan
NDLTD

NJIT's ETD project was given an ACRL/NJ Technology Innovation Honorable Mention Award in spring 2003