東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

High performance schedulers for netw...

Yang, Mei.

Linked to FindBook

Google Book

Amazon

博客來

High performance schedulers for network switches and routers.

Record Type:	Electronic resources : Monograph/item
Title/Author:	High performance schedulers for network switches and routers./
Author:	Yang, Mei.
Description:	160 p.
Notes:	Source: Dissertation Abstracts International, Volume: 64-07, Section: B, page: 3378.
Contained By:	Dissertation Abstracts International64-07B.
Subject:	Computer Science. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3098558

High performance schedulers for network switches and routers.
Yang, Mei.

High performance schedulers for network switches and routers. - 160 p.

Source: Dissertation Abstracts International, Volume: 64-07, Section: B, page: 3378.

Thesis (Ph.D.)--The University of Texas at Dallas, 2003.

This dissertation studies the cell scheduling problem for input queueing (IQ) and combined input and output queueing (CIOQ) switches with virtual output queues (VOQs) and focuses on constructing high performance schedulers for network switches and routers. To provide efficient and practical solutions, we take algorithm and hardware codesign as the theme of the dissertation. We propose several scheduler solutions, each consisting of scheduling algorithms, designs of special hardware components which can be used to build up schedulers based on these algorithms, and necessary switch architectures and queueing schemes.Subjects--Topical Terms:

626642
Computer Science.

High performance schedulers for network switches and routers.
LDR:03851nmm 2200289 4500 001 1857311
005 20040809115647.5
008 130614s2003 eng d
035 $a (UnM)AAI3098558
035 $a AAI3098558
040 $a UnM $c UnM
100 1 $a Yang, Mei. $3 1252651
245 1 0 $a High performance schedulers for network switches and routers.
300 $a 160 p.
500 $a Source: Dissertation Abstracts International, Volume: 64-07, Section: B, page: 3378.
500 $a Supervisor: S. Q. Zheng.
502 $a Thesis (Ph.D.)--The University of Texas at Dallas, 2003.
520 $a This dissertation studies the cell scheduling problem for input queueing (IQ) and combined input and output queueing (CIOQ) switches with virtual output queues (VOQs) and focuses on constructing high performance schedulers for network switches and routers. To provide efficient and practical solutions, we take algorithm and hardware codesign as the theme of the dissertation. We propose several scheduler solutions, each consisting of scheduling algorithms, designs of special hardware components which can be used to build up schedulers based on these algorithms, and necessary switch architectures and queueing schemes.
520 $a We summarize the results obtained in this dissertation as follows. (1) A pipelined framework for a class of pipelined iterative maximal size matching (MSM) algorithms, which reduce the scheduling time constraint by <math> <f> <fr><nu>SI</nu><de>S+I-1</de></fr></f> </math> times, where <italic>S</italic> is the speedup factor and <italic> I</italic> is the number of iterations allowed in each scheduling cycle. (2) A parallel round-robin arbiter (PRRA) design with <italic>O</italic>(log <italic> N</italic>)-gate delay and <italic>O</italic>(<italic>N</italic>) number of gates, which can be used to implement the proposed pipelined iterative MSM algorithms. (3) A new CIOQ switch architecture with space-division multiplexing expansion and grouped input/output ports (SDMG CIOQ switches for short), which only requires the switching fabric and memories to operate at the line rate. (4) Using fluid model techniques, we prove that any maximal size <italic> k</italic>-matching algorithm on an SDMG CIOQ switch with an expansion of 2 can achieve 100% throughput assuming input arrivals satisfy the strong law of large numbers (SLLN) and no input/output line is oversubscribed. (5) An efficient and starvation-free maximal size <italic>k</italic>-matching scheduling algorithm, the <italic>k</italic>-connection FIRM-based round-robin (<italic> k</italic>FRR) algorithm, for the SDMG CIOQ switch. (6) Programmable <italic> k</italic>-selector designs which are capable of selecting <italic>k</italic> out of <italic>N</italic> requests in <italic>O</italic>(log <italic>N</italic>)-gate delay, which can be used to implement the <italic>k</italic>FRR algorithm. (7) The dynamic DiffServ scheduling (DDS) algorithm for IQ switches, which provides minimum bandwidth guarantees for EF and AF traffic and fair bandwidth allocation for BE traffic. (8) Programmable weighted arbiter (PWA) designs with <italic>O</italic>(<italic>N</italic>) gates and <italic> O</italic>(<italic>b</italic> log <italic>N</italic>)-gate delay, where <italic> b</italic> is the number of bits needed to represent the weight. The PWA designs can be used to implement the DDS algorithm.
520 $a Performance of these algorithms and designs is evaluated by either proofs or simulations. Comparisons have been made with existing best solutions.
590 $a School code: 0382.
650 4 $a Computer Science. $3 626642
650 4 $a Engineering, Electronics and Electrical. $3 626636
690 $a 0984
690 $a 0544
710 2 0 $a The University of Texas at Dallas. $3 1018411
773 0 $t Dissertation Abstracts International $g 64-07B.
790 1 0 $a Zheng, S. Q., $e advisor
790 $a 0382
791 $a Ph.D.
792 $a 2003
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3098558