Merging Retrieval Requests for Multimedia Storage Server

Een Jun Hwang, B. Prabhakaran

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Most work on multimedia storage systems has assumed that clients will be serviced using a round-robin strategy. The server services the clients in rounds and each client is allocated a time slice within that round. Furthermore, most such algorithms are evaluated on the basis of a tightly specified cost function. This is the basis for well known algorithms such as FCFS, SCAN, SCAN-EDF, etc. In this paper, we first describe a request merging module that takes as input, a set of client requests, a set of constraints on the desired performance, and a cost function. It produces as output, a unified read request(URR), telling the storage server, which data items to read, and when these data items to be delivered to clients. And then, we present three algorithms, each of which accomplishes request merging. The first algorithm OptURR is guaranteed to produce optimal URRs with respect to arbitrary cost functions. Since the problem of computing an optimal URR is NP-complete, we develop two other algorithms that may produce sub-optimal URRs, but which have some nicer computational properties. We will report on the pros and cons of these algorithms through an experimental evaluation.

Original languageEnglish
Title of host publicationProceedings of the Joint Conference on Information Sciences
EditorsP.P. Wang, P.P. Wang
Pages616-619
Number of pages4
Volume5
Edition2
Publication statusPublished - 2000 Dec 1
Externally publishedYes
EventProceedings of the Fifth Joint Conference on Information Sciences, JCIS 2000 - Atlantic City, NJ, United States
Duration: 2000 Feb 272000 Mar 3

Other

OtherProceedings of the Fifth Joint Conference on Information Sciences, JCIS 2000
CountryUnited States
CityAtlantic City, NJ
Period00/2/2700/3/3

Fingerprint

Merging
Servers
Cost functions

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Hwang, E. J., & Prabhakaran, B. (2000). Merging Retrieval Requests for Multimedia Storage Server. In P. P. Wang, & P. P. Wang (Eds.), Proceedings of the Joint Conference on Information Sciences (2 ed., Vol. 5, pp. 616-619)

Merging Retrieval Requests for Multimedia Storage Server. / Hwang, Een Jun; Prabhakaran, B.

Proceedings of the Joint Conference on Information Sciences. ed. / P.P. Wang; P.P. Wang. Vol. 5 2. ed. 2000. p. 616-619.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hwang, EJ & Prabhakaran, B 2000, Merging Retrieval Requests for Multimedia Storage Server. in PP Wang & PP Wang (eds), Proceedings of the Joint Conference on Information Sciences. 2 edn, vol. 5, pp. 616-619, Proceedings of the Fifth Joint Conference on Information Sciences, JCIS 2000, Atlantic City, NJ, United States, 00/2/27.
Hwang EJ, Prabhakaran B. Merging Retrieval Requests for Multimedia Storage Server. In Wang PP, Wang PP, editors, Proceedings of the Joint Conference on Information Sciences. 2 ed. Vol. 5. 2000. p. 616-619
Hwang, Een Jun ; Prabhakaran, B. / Merging Retrieval Requests for Multimedia Storage Server. Proceedings of the Joint Conference on Information Sciences. editor / P.P. Wang ; P.P. Wang. Vol. 5 2. ed. 2000. pp. 616-619
@inproceedings{ff0543286a304ab7b74ace23c18caa97,
title = "Merging Retrieval Requests for Multimedia Storage Server",
abstract = "Most work on multimedia storage systems has assumed that clients will be serviced using a round-robin strategy. The server services the clients in rounds and each client is allocated a time slice within that round. Furthermore, most such algorithms are evaluated on the basis of a tightly specified cost function. This is the basis for well known algorithms such as FCFS, SCAN, SCAN-EDF, etc. In this paper, we first describe a request merging module that takes as input, a set of client requests, a set of constraints on the desired performance, and a cost function. It produces as output, a unified read request(URR), telling the storage server, which data items to read, and when these data items to be delivered to clients. And then, we present three algorithms, each of which accomplishes request merging. The first algorithm OptURR is guaranteed to produce optimal URRs with respect to arbitrary cost functions. Since the problem of computing an optimal URR is NP-complete, we develop two other algorithms that may produce sub-optimal URRs, but which have some nicer computational properties. We will report on the pros and cons of these algorithms through an experimental evaluation.",
author = "Hwang, {Een Jun} and B. Prabhakaran",
year = "2000",
month = "12",
day = "1",
language = "English",
isbn = "0964345692",
volume = "5",
pages = "616--619",
editor = "P.P. Wang and P.P. Wang",
booktitle = "Proceedings of the Joint Conference on Information Sciences",
edition = "2",

}

TY - GEN

T1 - Merging Retrieval Requests for Multimedia Storage Server

AU - Hwang, Een Jun

AU - Prabhakaran, B.

PY - 2000/12/1

Y1 - 2000/12/1

N2 - Most work on multimedia storage systems has assumed that clients will be serviced using a round-robin strategy. The server services the clients in rounds and each client is allocated a time slice within that round. Furthermore, most such algorithms are evaluated on the basis of a tightly specified cost function. This is the basis for well known algorithms such as FCFS, SCAN, SCAN-EDF, etc. In this paper, we first describe a request merging module that takes as input, a set of client requests, a set of constraints on the desired performance, and a cost function. It produces as output, a unified read request(URR), telling the storage server, which data items to read, and when these data items to be delivered to clients. And then, we present three algorithms, each of which accomplishes request merging. The first algorithm OptURR is guaranteed to produce optimal URRs with respect to arbitrary cost functions. Since the problem of computing an optimal URR is NP-complete, we develop two other algorithms that may produce sub-optimal URRs, but which have some nicer computational properties. We will report on the pros and cons of these algorithms through an experimental evaluation.

AB - Most work on multimedia storage systems has assumed that clients will be serviced using a round-robin strategy. The server services the clients in rounds and each client is allocated a time slice within that round. Furthermore, most such algorithms are evaluated on the basis of a tightly specified cost function. This is the basis for well known algorithms such as FCFS, SCAN, SCAN-EDF, etc. In this paper, we first describe a request merging module that takes as input, a set of client requests, a set of constraints on the desired performance, and a cost function. It produces as output, a unified read request(URR), telling the storage server, which data items to read, and when these data items to be delivered to clients. And then, we present three algorithms, each of which accomplishes request merging. The first algorithm OptURR is guaranteed to produce optimal URRs with respect to arbitrary cost functions. Since the problem of computing an optimal URR is NP-complete, we develop two other algorithms that may produce sub-optimal URRs, but which have some nicer computational properties. We will report on the pros and cons of these algorithms through an experimental evaluation.

UR - http://www.scopus.com/inward/record.url?scp=1642314272&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1642314272&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:1642314272

SN - 0964345692

VL - 5

SP - 616

EP - 619

BT - Proceedings of the Joint Conference on Information Sciences

A2 - Wang, P.P.

A2 - Wang, P.P.

ER -