Distributed Resource Control

Microsoft Library Research

Researcher: Pam Cahn

Prepared for: Lakshminarayanan Subramanian

Date: 8/3/00

Research Time: 1 hour

Data Costs: $75

Subject: Resource Allocation, Agents, and Brokers

Direct costs and amount of time spent for this research are provided for your information and are not being charged to your department's budget.

This material is copyrighted and should not be distributed outside of Microsoft.

______________________________________________________

To request a copy of any abstracted article or paper, e-mail the citation directly to the “library” alias.

6549099 INSPEC Abstract Number: B2000-05-6150P-010, C2000-05-5630-006

Title: RATES: a server for MPLS traffic engineering

Author(s): Aukia, P.; Kodialam, M.; Koppol, P.V.N.; Lakshman, T.V.;

Sarin, H.; Suter, B.

Author Affiliation: Lucent Technol., AT&T Bell Labs., USA

Journal: IEEE Network vol.14, no.2 p.34-41

Publisher: IEEE,

Publication Date: March-April 2000 Country of Publication: USA

CODEN: IENEET ISSN: 0890-8044

SICI: 0890-8044(200003/04)14:2L.34:RSMT;1-4

Material Identity Number: J991-2000-002

U.S. Copyright Clearance Center Code: 0890-8044/2000/$10.00

Language: English Document Type: Journal Paper (JP)

Treatment: Applications (A); New Developments (N); Practical (P)

Abstract: It has been suggested that one of the most significant reasons

for multiprotocol label switching (MPLS) network deployment is network

traffic engineering. The goal of traffic engineering is to make the best

use of the network infrastructure, and this is facilitates by the explicit

routing feature of MPLS, which allows many of the shortcomings associated

with current IP routing schemes to be addressed. This article describes a

software system called Routing and Traffic Engineering Server (RATES)

developed for MPLS traffic engineering. It also describes some new routing

ideas incorporated in RATES for MPLS explicit path selection. The RATES

implementation consists of a policy and flow database, a browser-based

interface for policy definition and entering resource provisioning

requests, and a Common Open Policy Service protocol server-client

implementation for communicating paths and resource information to edge

routers. RATES also uses the OSPF topology database for dynamically

obtaining link state information. RATES can set up bandwidth-guaranteed

label-switched (LSPs) between specified ingress-egress pairs. The path

selection for LSPs is on a new minimum-interference routing algorithm aimed

at making the best use of network infrastructure in an online environment

where LSP requests arrive one by one with no a priori information about

future requests. Although developed for an MPLS application, the RATES

implementation has many similarities in components to an intradomain

differentiated services bandwidth broker. (21 Refs)

Descriptors: distributed databases; network servers; network topology;

online operation; open systems; packet switching; telecommunication network

routing; telecommunication traffic; transport protocols

Identifiers: MPLS traffic engineering; RATES; server; multiprotocol label

switching; MPLS network deployment; network traffic engineering; network

infrastructure; IP routing; software system; Routing and Traffic

Engineering Server; path selection; policy database; flow database;

browser-based interface; policy definition; resource provisioning requests;

Common Open Policy Service; protocol server-client implementation; edge

routers; resource information; OSPF topology database; link state

information; path information; bandwidth-guaranteed label-switched;

minimum-interference routing algorithm; intradomain differentiated services

bandwidth broker

Class Codes: B6150P (Communication network design, planning and routing);

B6210L (Computer communications); B6150M (Protocols); B6150C (Communication

switching); C5630 (Networking equipment); C5640 (Protocols); C6150N (

Distributed systems software); C6160B (Distributed databases)

6635671 INSPEC Abstract Number: B2000-08-6210L-071, C2000-08-5620W-017

Title: A two-tier resource management model for the Internet

Author(s): Terzis, A.; Wang, L.; Ogawa, J.; Zhang, L.

Author Affiliation: Dept. of Comput. Sci., California Univ., Los Angeles,

CA, USA

Conference Title: Seamless Interconnection for Universal Services. Global

Telecommunications Conference. GLOBECOM'99. (Cat. No.99CH37042) Part

vol.3 p.1779-91 vol.3

Publisher: IEEE, Piscataway, NJ, USA

Publication Date: 1999 Country of Publication: USA 6 vol.(lii+2798)

pp.

ISBN: 0 7803 5796 5 Material Identity Number: XX-2000-00490

U.S. Copyright Clearance Center Code: 0 7803 5796 5/99/$10.00

Conference Title: Seamless Interconection for Universal Services. Global

Telecommunications Conference. GLOBECOM'99

Conference Date: 5-9 Dec. 1999 Conference Location: Rio de Janeireo,

Brazil

Language: English Document Type: Conference Paper (PA)

Treatment: Theoretical (T)

Abstract: In this paper we propose a two-tier resource management model

for the global Internet. Our solution resembles the current two-tier

routing hierarchy and allows individual administrative domains to

independently make their own decisions on strategies and protocols to use

for internal resource management and QoS support. The aggregate traffic

crossing domain borders is served according to relatively stable,

long-lived bilateral agreements. End-to-end QoS support is achieved through

the concatenation of such bilateral agreements. We describe in detail a

realization of this two-tier model, where a bandwidth broker (BB) acts as

the resource manager for each administrative domain, neighboring bandwidth

brokers communicate with each other to establish inter-domain resource

agreements. As an illustrative example in this paper we used a simplified

RSVP as an intradomain resource allocation protocol for the aggregate

traffic between border routers. Our simulation results show that this

two-tier design can provide effective end-to-end QoS support for user

applications. (17 Refs)

Descriptors: computer network management; Internet; protocols; quality of

service

Identifiers: two-tier resource management model; Internet; administrative

domains; strategies; protocols; internal resource management; QoS support;

aggregate traffic; bilateral agreements; bandwidth broker; neighboring

bandwidth brokers; inter-domain resource agreements; RSVP; intradomain

resource allocation protocol

Class Codes: B6210L (Computer communications); B6150M (Protocols); B6210C

(Network management); C5620W (Other computer networks); C5640 (Protocols)

6529366 INSPEC Abstract Number: B2000-04-6210L-092, C2000-04-5620W-033

Title: Evaluation of bandwidth broker signaling

Author(s): Gunter, M.; Braun, T.

Author Affiliation: Inst. of Comput. Sci. & Appl. Math., Bern Univ.,

Switzerland

Conference Title: Proceedings Seventh International Conference on Network

Protocols (ICNP'99) p.145-52

Publisher: IEEE Comput. Soc, Los Alamitos, CA, USA

Publication Date: 1999 Country of Publication: USA xiii+345 pp.

ISBN: 0 7695 0412 4 Material Identity Number: XX-1999-03051

U.S. Copyright Clearance Center Code: 0 7695 0412 4/99/$10.00

Conference Title: Proceedings of ICNP'99: 7th International Conference on

Network Protocols

Conference Sponsor: IEEE Comput. Soc. Tech. Committee on Distributed

Process

Conference Date: 31 Oct.-3 Nov. 1999 Conference Location: Toronto,

Ont., Canada

Language: English Document Type: Conference Paper (PA)

Treatment: Applications (A); Practical (P); Theoretical (T)

Abstract: The differentiated services (DiffServ) architecture for the

Internet implements a scalable mechanism for quality-of-service (QoS)

provisioning. Bandwidth brokers represent the instances of the

architecture, that automate the provisioning of a DiffServ service between

network domains. Although several bandwidth broker implementations have

been proposed, the alternatives and trade-offs of the different viable

approaches of inter-broker communication were not studied up to now. This

paper presents the broker signaling trade-offs considered in the context of

a DiffServ scenario used by the Swiss National Science Foundation project

CATI, and it presents results gathered by simulations. (10 Refs)

Descriptors: Internet; quality of service; telecommunication signalling;

telecommunication traffic

Identifiers: bandwidth broker signaling; differentiated services

architecture; Internet; scalable mechanism; quality of service provisioning

; QoS; network domains; inter-broker communication; Swiss National Science

Foundation; CATI project

Class Codes: B6210L (Computer communications); C5620W (Other computer

networks); C7210N (Information networks)

6521061 INSPEC Abstract Number: B2000-04-6250F-294

Title: Hybrid channel assignment using dynamic resource sharing

Author(s): Al Agha, K.

Author Affiliation: PRiSM Lab., Versailles Univ., France

Conference Title: Gateway to 21st Century Communications Village. VTC

1999-Fall. IEEE VTS 50th Vehicular Technology Conference (Cat.

No.99CH36324) Part vol.5 p.3029-33 vol.5

Publisher: IEEE, Piscataway, NJ, USA

Publication Date: 1999 Country of Publication: USA 5 vol. (lxix+3056)

pp.

ISBN: 0 7803 5435 4 Material Identity Number: XX-1999-02881

U.S. Copyright Clearance Center Code: 0 7803 5435 4/99/$10.00

Conference Title: Gateway to 21st Century Communications Village. VTC

1999-Fall. IEEE VTS 50th Vehicular Technology Conference

Conference Date: 19-22 Sept. 1999 Conference Location: Amsterdam,

Netherlands

Language: English Document Type: Conference Paper (PA)

Treatment: Practical (P)

Abstract: Resource allocation in wireless networks needs to be adaptive

in order to cope with non-uniform traffic variation on an hourly and daily

basis. In all future systems, the use of dynamic and adaptive resource

assignment will become a necessity in order to optimize the system

capacity. In this context, the use of multi-agent systems represents a

potential solution to helping the assignment process. In addition,

communication between agents adds more information on the cell environment

and reduces conflicts between neighboring base stations in order to

allocate the same channel. In this paper, we propose a hybrid channel

assignment with a dynamic sharing between the fixed and dynamic part of the

resource common pool. (12 Refs)

Descriptors: channel allocation; channel capacity; mobile radio;

multi-agent systems; optimisation; packet radio networks; telecommunication

traffic

Identifiers: hybrid channel assignment; dynamic resource sharing;

wireless networks; non-uniform traffic variation; adaptive resource

assignment; optimization; system capacity; multi-agent systems; cell

environment; conflict reduction; base stations

Class Codes: B6250F (Mobile radio systems); B0260 (Optimisation

techniques)

6513558 INSPEC Abstract Number: C2000-04-6150N-022

Title: Communication infrastructure of scheduling agents

Author(s): Yen, B.P.-C.; Chow, K.-W.

Author Affiliation: Dept. of Ind. Eng. & Eng. Manage., Hong Kong Univ. of

Sci. & Technol., Hong Kong

Conference Title: IEEE SMC'99 Conference Proceedings. 1999 IEEE

International Conference on Systems, Man, and Cybernetics (Cat.

No.99CH37028) Part vol.4 p.552-7 vol.4

Publisher: IEEE, Piscataway, NJ, USA

Publication Date: 1999 Country of Publication: USA 6 vol.

(1179+1075+1106+1124+1140+1078) pp.

ISBN: 0 7803 5731 0 Material Identity Number: XX-1999-03291

U.S. Copyright Clearance Center Code: 0 7803 5731 0/99/$10.00

Conference Title: IEEE SMC'99 Conference Proceedings. 1999 IEEE

International Conference on Systems, Man, and Cybernetics

Conference Sponsor: IEEE Syst., Man, & Cybernetics Soc. (SMC); Sci.

Council of Japan (SCJ); Soc. Instrum. & Control Eng. (SICE); Robotics Soc.

Japan (RSJ); Japan Soc. Mech. Eng. (JSME)

Conference Date: 12-15 Oct. 1999 Conference Location: Tokyo, Japan

Language: English Document Type: Conference Paper (PA)

Treatment: Practical (P)

Abstract: Scheduling involves the allocation of limited resources to

tasks over time. Due to the high complexity of scheduling problems, most

scheduling algorithms and procedures are only suitable for specific cases

and most scheduling systems are very application-oriented. Until now, no

single scheduling system has been able to handle all types of scheduling

problems. With the development of the client/server architecture and

Internet technology, distributed problem solving by a group of systems

becomes feasible where systems can exchange information, share resources

and cooperate to solve large and complicated problems. Moreover, intensive

research into intelligent software agents and agent technology has offered

tools for solving problems and coordination between systems. These tools

allow systems to be represented as agents and coordinate their activities

through message passing. However, most of the concepts and approaches have

been applied to general applications and very little research has been

studied the distributed approach to scheduling systems development. In this

research, the framework of agent-based distributed scheduling is presented.

The research is divided into three parts: Internet scheduling systems; the

establishment of communication infrastructure; and the distributed problem

solving policy. (14 Refs)

Descriptors: client-server systems; Internet; message passing; problem

solving; resource allocation; software agents

Identifiers: communication infrastructure; scheduling agents; limited

resources; client/server architecture; Internet technology; distributed

problem; intelligent software agents; agent technology; agent-based

distributed scheduling; Internet scheduling systems; distributed problem

solving policy

Class Codes: C6150N (Distributed systems software); C6170K (Knowledge

engineering techniques); C7100 (Business and administration)

6376784 INSPEC Abstract Number: C1999-11-5440-017

Title: Resource allocation and scheduling in metasystems

Author(s): Schwiegelshohn, U.; Yahyapour, R.

Author Affiliation: Inst. of Comput. Eng., Dortmund Univ., Germany

Conference Title: High-Performance Computing and Networking. 7th

International Conference, HPCN Europe 1999. Proceedings p.851-60

Editor(s): Sloot, P.; Bubak, M.; Hoekstra, A.; Hertzberger, B.

Publisher: Springer-Verlag, Berlin, Germany

Publication Date: 1999 Country of Publication: Germany xxiii+1318 pp.

ISBN: 3 540 65821 1 Material Identity Number: XX-1999-02493

Conference Title: High-Performance Computing and Networking. 7th

International Conference, HPCN Europe 1999. Proceedings

Conference Date: 12-14 April 1999 Conference Location: Amsterdam,

Netherlands

Language: English Document Type: Conference Paper (PA)

Treatment: Practical (P)

Abstract: In this paper we present NWIRE, a new management architecture

for metacomputing systems. After the general introduction of the

architecture we first describe the properties that are relevant for

scheduling. Then we derive general requirements for scheduling strategies

in metasystems and point out differences to conventional job scheduling for

parallel processors. This leads to the metacomputing scheduling concept of

NWIRE which is based on a brokerage and trading approach. (4 Refs)

Descriptors: parallel architectures; processor scheduling; resource

allocation

Identifiers: resource allocation; scheduling; metasystems; NWIRE;

management architecture; metacomputing system; brokerage and trading

approach

Class Codes: C5440 (Multiprocessing systems); C6150N (Distributed

systems software); C5220P (Parallel architecture)

6365325 INSPEC Abstract Number: C1999-11-6150N-028

Title: Locust-a brokerage system for accessing idle resources for

Web-Computing

Author(s): May, M.

Author Affiliation: Dept. of Comput. Sci., Tech. Univ. Munchen, Germany

Conference Title: Proceedings 25th EUROMICRO Conference. Informatics:

Theory and Practice for the New Millennium Part vol.2 p.466-73 vol.2

Publisher: IEEE Comput. Soc, Los Alamitos, CA, USA

Publication Date: 1999 Country of Publication: USA 2 vol.

(xxviii+530+478) pp.

ISBN: 0 7695 0321 7 Material Identity Number: XX-1999-02506

U.S. Copyright Clearance Center Code: 1089 6503/99/$10.00

Conference Title: Proceedings of Euromicro Workshop

Conference Date: 8-10 Sept. 1999 Conference Location: Milan, Italy

Language: English Document Type: Conference Paper (PA)

Treatment: Practical (P)

Abstract: In this paper we present a brokerage system for making idle cpu

resources of anonymous users of the World Wide Web accessible for

distributed Web-Computing. The Locust (LOw cost Computing Utilizing Skimmed

idle Time) Broker collects and remarkets the aggregated resources by means

of ubiquitous Web and Java technologies and creates an electronic spot

market focused on the mainstream Internet user as suppliers of idle

computing resources. We introduce the economic Price/Market model of Locust

enabling the export or import of resource surpluses or lacks between

so-called submarkets. Test cases for consumption of aggregated CPU

resources include distributed Raytracing and RC5 decryption. We further

describe the RC5 key block management and our efforts to reach performance

comparable to native RC5 implementations. (18 Refs)

Descriptors: resource allocation; wide area networks

Identifiers: Locust; brokerage system; idle resources; distributed

Web-Computing; aggregated CPU resources

Class Codes: C6150N (Distributed systems software); C5620W (Other

computer networks)

6131057 INSPEC Abstract Number: B1999-02-6210L-085, C1999-02-5620-022

Title: Specifying resources and services in metacomputing environments

Author(s): Brune, M.; Gehring, J.; Keller, A.; Monien, B.; Ramme, F.;

Reinefeld, A.

Author Affiliation: Centre for Parallel Comput., Paderborn Univ., Germany

Journal: Parallel Computing vol.24, no.12-13 p.1751-76

Publisher: Elsevier,

Publication Date: Nov. 1998 Country of Publication: Netherlands

CODEN: PACOEJ ISSN: 0167-8191

SICI: 0167-8191(199811)24:12/13L.1751:SRSM;1-U

Material Identity Number: F777-1998-012

U.S. Copyright Clearance Center Code: 0167-8191/98/$19.00

Document Number: S0167-8191(98)00076-3

Language: English Document Type: Journal Paper (JP)

Treatment: Applications (A); Practical (P)

Abstract: With a steadily increasing number of services, metacomputing is

now gaining importance in science and industry. Virtual organizations,

autonomous agents, mobile computing services, and high-performance

client-server applications are among the many examples of metacomputing

services. For all of them, resource description plays a major role in

organizing access, use, and administration of the computing components and

software services. We present a generic Resource and Service Description

(RSD) for specifying the hardware and software components of (meta-)

computing environments. Its graphical interface allows metacomputer users

to specify their resource requests. Its textual counterpart gives service

providers the necessary flexibility to specify topology and properties of

the available system and software resources, Finally, its internal

object-oriented representation is used to link different resource

management systems and service tools. With these three representations, our

generic RSD approach is a key component for building metacomputer

environments. (48 Refs)

Descriptors: computer networks; Internet; mobile computing; programming

environments; resource allocation

Identifiers: metacomputing environments; virtual organizations;

autonomous agents; mobile computing services; high-performance

client-server applications; resource description; software services;

resource and service description; graphical interface; object-oriented

representation

Class Codes: B6210L (Computer communications); C5620 (Computer networks

and techniques); C6150J (Operating systems); C6150N (Distributed systems

software); C6115 (Programming support); C7210N (Information networks)

6439045 INSPEC Abstract Number: B2000-01-6210L-168, C2000-01-5620-048

Title: Specifying resources and services in metacomputing systems

Author(s): Brune, M.; Gehring, J.; Keller, A.; Reinefeld, A.

Author Affiliation: Konrad-Zuse-Zentrum fur Inf., Berlin, Germany

Book Title: High performance cluster computing: Architectures and

systems, Vol. 1 p.186-200

Editor(s): Buyya, R.

Publisher: Prentice Hall PTR, Upper Saddle River, NJ, USA

Publication Date: 1999 Country of Publication: USA 2

vol.(xxxiii+849+xxiv+664) pp.

ISBN: 0 13 013784 7 Material Identity Number: XB-1999-00101

Language: English Document Type: Book Chapter (BC)

Treatment: Practical (P)

Abstract: With the metacomputing paradigm getting more and more accepted,

an increasing number of network based services is offered. Today, the

available services range from traditional client-server applications over

distributed intra/Internet services, virtual organizations, teleworking,

teleconferencing, up to autonomous agents for information gathering

purposes. All these services are coordinated by a metacomputing middleware,

which is also geographically distributed. Clearly, this middleware must be

generic and versatile to also serve newly emerging technologies. We present

a scheme for specifying resources and services in complex heterogeneous

computing systems and metacomputing environments. The scheme is named RSD

(Resource and Services Description). At the system administrator level, RSD

is used to specify the available system components, such as the number of

nodes, their interconnection topology, CPU speeds, and available software

packages. At the user level, a GUI provides a comfortable, high-level

interface for specifying system requests. A textual editor can be used for

defining repetitive and recursive structures. This gives service providers

the necessary flexibility for fine-grained specification of system

topologies, interconnection networks, and system and software dependent

properties. All these representations are mapped onto a single, coherent

internal object oriented resource representation. Dynamic aspects are

traced at runtime and included in the resource description to allow for

optimal process mapping and dynamical task load balancing at runtime at the

metacomputer level. This is done in a self-organizing way, with human

system operators becoming involved only when new hardware/software

components are installed. (16 Refs)

Descriptors: client-server systems; computer networks; graphical user

interfaces; interactive systems; object-oriented programming; performance

evaluation; resource allocation; text editing; workstation clusters

Identifiers: metacomputing systems; metacomputing paradigm; network based

services; traditional client-server applications; distributed

intra/Internet services; virtual organizations; teleworking;

teleconferencing; autonomous agent; information gathering; metacomputing

middleware; resource specification; complex heterogeneous computing systems

; metacomputing environments; RSD; Resource and Services Description;

system administrator level; interconnection topology; CPU speeds; software

packages; high-level interface; system requests; textual editor; recursive

structures; service providers; fine-grained specification; system

topologies; interconnection networks; software dependent properties;

coherent internal object oriented resource representation; dynamic aspects;

resource description; optimal process mapping; dynamical task load

balancing; metacomputer level; human system operators; hardware/software

components

Class Codes: B6210L (Computer communications); C5620 (Computer networks

and techniques); C6150N (Distributed systems software); C6180G (Graphical

user interfaces); C6130 (Data handling techniques); C6110J (

Object-oriented programming); C5470 (Performance evaluation and testing);

C5670 (Network performance)

6409751 INSPEC Abstract Number: C1999-12-6180-021

Title: Using Gateway system to provide a desktop access to high

performance computational resources

Author(s): Akarsu, E.; Fox, G.; Haupt, T.; Kalinichenko, A.; Kang-Seok

Kim; Sheethaalnath, P.; Choon-Han Youn

Author Affiliation: Northeast Parallel Archit. Center, Syracuse Univ.,

NY, USA

Conference Title: Proceedings. The Eighth International Symposium on High

Performance Distributed Computing (Cat. No.99TH8469) p.294-8

Publisher: IEEE Comput. Soc, Los Alamitos, CA, USA

Publication Date: 1999 Country of Publication: USA xii+359 pp.

ISBN: 0 7803 5681 0 Material Identity Number: XX-1999-02913

U.S. Copyright Clearance Center Code: 0 7803 5681 0/99/$10.00

Conference Title: Proceedings. The Eighth International Symposium on High

Performance Distributed Computing

Conference Sponsor: IEEE; Univ. Southern California; Univ. California;

Nat. Partnership for Adv. Computational Infrastructure; Univ. Arizona;

Microsoft; Sun Microsyst. Compuuter Co

Conference Date: 3-6 Aug. 1999 Conference Location: Redondo Beach, CA,

USA

Language: English Document Type: Conference Paper (PA)

Treatment: Practical (P)

Abstract: In this paper, we discuss the use of Gateway for seamless

desktop access to high performance resources. We illustrate our ideas with

two Gateway applications that require access to remote resources: the

Landscape Management System (LMS) and Quantum Simulations (QS). For LMS we

use Gateway to retrieve data from many different sources as well as to

allocate remote computational resources needed to solve the problem at

hand. Gateway transparently controls the necessary data transfer between

hosts for the user. Quantum Simulations requires access to HPCC resources

and therefore we layered Gateway on top of the Globus metacomputing

toolkit. This way Gateway plays the role of a job broker for Globus. (17

Refs)

Descriptors: client-server systems; online front-ends; resource

allocation; user interfaces

Identifiers: Gateway system; desktop access; high performance

computational resources; remote resource access; Landscape Management

System; Quantum Simulations; data retrieval; remote computational resource

allocation; data transfer control; hosts; Globus metacomputing toolkit; job

broker

Class Codes: C6180 (User interfaces); C7250N (Search engines); C6150N (

Distributed systems software)

6513510 INSPEC Abstract Number: C2000-04-7400-003

Title: Simulation of a multi-agent protocol for task allocation in

cooperative design

Author(s): Kwang Mong Sim; Shiu, S.C.K.; Bun, M.L.

Author Affiliation: Dept. of Comput., Hong Kong Polytech., Kowloon, Hong

Kong

Conference Title: IEEE SMC'99 Conference Proceedings. 1999 IEEE

International Conference on Systems, Man, and Cybernetics (Cat.

No.99CH37028) Part vol.3 p.95-100 vol.3

Publisher: IEEE, Piscataway, NJ, USA

Publication Date: 1999 Country of Publication: USA 6 vol.

(1179+1075+1106+1124+1140+1078) pp.

ISBN: 0 7803 5731 0 Material Identity Number: XX-1999-03290

U.S. Copyright Clearance Center Code: 0 7803 5731 0/99/$10.00

Conference Title: IEEE SMC'99 Conference Proceedings. 1999 IEEE

International Conference on Systems, Man, and Cybernetics

Conference Sponsor: IEEE Syst., Man, & Cybernetics Soc. (SMC); Sci.

Council of Japan (SCJ); Soc. Instrum. & Control Eng. (SICE); Robotics Soc.

Japan (RSJ); Japan Soc. Mech. Eng. (JSME)

Conference Date: 12-15 Oct. 1999 Conference Location: Tokyo, Japan

Language: English Document Type: Conference Paper (PA)

Treatment: Applications (A); Practical (P); Theoretical (T)

Abstract: To achieve high performance in agent-based cooperative design,

the effective allocation of design agents to distributed and cooperative

design tasks becomes a crucial issue. This research extends the well-known

contract net protocol for decentralized task allocation in cooperative

engineering design. Since the contract net protocol only provides a generic

framework for agents to exchange and evaluate information, it does not

prescribe any specific coordination policies for cooperative engineering

design. By adding a set of agent selection and task selection policies,

this research addresses issues in agent-based cooperative engineering

design such as: How is an agent selected to carry out a design task? How

can an agent select from among a list of design tasks which to bid for?

Through a series of colored Petri-net simulation experiments, the

performance of these selection policies are measured and evaluated. Smith

(1980) noted that the contract net protocol is particularly suitable for

executing different (sub-)tasks simultaneously while ensuring effective

resource allocation and focused decision. Results from the coloured

Petri-net simulations have shown that these properties are preserved in the

proposed protocol. (11 Refs)

Descriptors: CAD; graph colouring; multi-agent systems; Petri nets;

resource allocation; software agents

Identifiers: multi-agent protocol; task allocation; agent-based

cooperative design; design agents; cooperative engineering design; agent

selection; task selection; colored Petri-net simulation

Class Codes: C7400 (Engineering computing); C6170 (Expert systems and

other AI software and techniques); C1160 (Combinatorial mathematics);

C6150J (Operating systems)

6090835 INSPEC Abstract Number: C9901-6170K-004

Title: Adaptive choice of information sources

Author(s): Sen, S.

Author Affiliation: Dept. of Math. & Comput. Sci., Tulsa Univ., OK, USA

Conference Title: Cooperative Information Agents II. Learning, Mobility

and Electronic Commerce for Information Discovery on the Internet. Second

International Workshop, CIA'98. Proceedings p.190-4

Editor(s): Klusch, M.; Weiss, G.

Publisher: Springer-Verlag, Berlin, Germany

Publication Date: 1998 Country of Publication: Germany ix+308 pp.

ISBN: 3 540 64676 0 Material Identity Number: XX98-01825

Conference Title: Cooperative Information Agents II Learning, Mobility

and Electronic Commerce for Information Discovery on the Internet. Second

International Workshop, CIA'98. Proceedings

Conference Date: 4-7 July 1998 Conference Location: Paris, France

Language: English Document Type: Conference Paper (PA)

Treatment: Practical (P)

Abstract: We present a number of learning approaches by which agents can

adapt to select information sources that satisfy performance requirements.

Performance can be interpreted both in terms of the quality of information

provided by the sources, as well as the response time to process

information requests. We first present a couple of approaches by which

self-motivated agents can learn to choose lightly-loaded resources. The

resultant load balancing effect results in increasing throughput for the

entire system as well as faster response times for individual agents. We

also present an expected utility maximization approach to selecting

information sources that are likely to deliver better quality information

to different classes of queries. (0 Refs)

Descriptors: cooperative systems; learning (artificial intelligence);

resource allocation; software agents; software performance evaluation

Identifiers: information sources; learning; agents; performance

requirements; information quality; response time; self-motivated agents;

load balancing; expected utility maximization; queries; adaptive choice

Class Codes: C6170K (Knowledge engineering techniques); C1240 (Adaptive

system theory); C1230 (Artificial intelligence)

6076098 INSPEC Abstract Number: B9812-6150M-051, C9812-5640-049

Title: Cooperation protocols for multimedia communication network based on

ADIPS framework

Author(s): Suganuma, T.; Kinoshita, T.; Sugawara, K.; Shiratori, N.

Author Affiliation: Res. Inst. of Electr. Commun., Tohoku Univ., Sendai,

Japan

Conference Title: Proceedings of the 1998 ICPP Workshop on Architectural

and OS Support for Multimedia Applications Flexible Communication Systems.

Wireless Networks and Mobile Computing (Cat. No.98EX206) p.76-85

Editor(s): Das, C.R.

Publisher: IEEE Comput. Soc, Los Alamitos, CA, USA

Publication Date: 1998 Country of Publication: USA vii+155 pp.

ISBN: 0 8186 8657 X Material Identity Number: XX98-02123

U.S. Copyright Clearance Center Code: 0 8186 8657 X/98/$10.00

Conference Title: Proceeding of Workshop on Architectural and Operation

Systems Support for Multimedia Applications

Conference Sponsor: Int. Assoc. Comput. & Commun.; Ohio State Univ

Conference Date: 14 Aug. 1998 Conference Location: Minneapolis, MN,

USA

Language: English Document Type: Conference Paper (PA)

Treatment: Practical (P)

Abstract: The growth of both computer communication such as the Internet,

and the computational power of computers makes multimedia applications

possible for common use. However, some significant problems still remained

unsolved, e.g., the efficient use of resources and the dynamic control of

applications for satisfying the users' required QoS. To solve these

problems, we propose an agent based approach to realize flexible multimedia

communication based on the ADIPS Framework. We explain the cooperation

protocols used in our flexible video conference system and also demonstrate

the flexibility in terms of QoS control achieved by agent cooperation. (18

Refs)

Descriptors: Internet; multimedia systems; protocols; quality control;

resource allocation; teleconferencing

Identifiers: cooperation protocols; multimedia communication network;

ADIPS framework; computer communication; Internet; computational power;

multimedia applications; dynamic control; QoS; agent based approach;

flexible multimedia communication; flexible video conference system; QoS

control; agent cooperation

Class Codes: B6150M (Protocols); B6210R (Multimedia communications);

B6210L (Computer communications); B6210P (Teleconferencing); C5640 (

Protocols); C6130M (Multimedia); C5620W (Other computer networks); C7210 (

Information services and centres); C7410F (Communications computing)

5992817 INSPEC Abstract Number: C9809-7100-040

Title: Multi-agent technology for planning, scheduling, and resource

allocation

Author(s): Gorodetski, V.; Lebedev, A.

Author Affiliation: St. Petersburg Inst. for Inf. & Autom., Russia

Conference Title: Proceedings International Conference on Multi Agent

Systems (Cat. No.98EX160) p.429-30

Publisher: IEEE Comput. Soc, Los Alamitos, CA, USA

Publication Date: 1998 Country of Publication: USA xviii+487 pp.

ISBN: 0 8186 8500 X Material Identity Number: XX98-01978

U.S. Copyright Clearance Center Code: 0 8186 8500 X/98/$110.00

Conference Title: Proceedings International Conference on Multi Agent

Systems

Conference Date: 3-7 July 1998 Conference Location: Paris, France

Language: English Document Type: Conference Paper (PA)

Treatment: Practical (P)

Abstract: A large number of hot applications are reducible to

combinatorial models. Almost all of them are at least NP-hard and cannot be

solved in a traditional way. We consider an agent-based approach to solve a

set of planning and scheduling applications under real time and temporal

constraints. The problem is formalised as a contract allocation task over a

set of companies and specified as an auction-based competition of agents

managed by a meta-agent. The contributions are: the multi-agent system

architecture; knowledge-based specification of real time and temporal

constraints; and a knowledge-based auction management model. (6 Refs)

Descriptors: business data processing; contracts; cooperative systems;

planning (artificial intelligence); real-time systems; resource allocation;

scheduling; software agents

Identifiers: multi-agent systems; planning; scheduling; resource

allocation; combinatorial models; NP-hard; real time constraints; temporal

constraints; contract allocation task; auction-based competition;

meta-agent; system architecture; knowledge-based specification;

knowledge-based auction management

Class Codes: C7100 (Business and administration); C6170 (Expert systems

); C1230 (Artificial intelligence)

6288285 INSPEC Abstract Number: C1999-08-7460-017

Title: Resource allocation in a distributed computing environment

Author(s): Carter, R.L.; St. Louis, D.; Andert, E.P., Jr.

Author Affiliation: Conceptual Syst. & Software, Chino Valley, AZ, USA

Conference Title: 17th DASC. AIAA/IEEE/SAE. Digital Avionics Systems

Conference. Proceedings (Cat. No.98CH36267) Part vol.1 p. C32/1-8

vol.1

Publisher: IEEE, New York, NY, USA

Publication Date: 1998 Country of Publication: USA 2 vol.(xi+xv+1444)

pp.

ISBN: 0 7803 5086 3 Material Identity Number: XX-1998-03217

U.S. Copyright Clearance Center Code: 0 7803 5086 3/98/$10.00

Conference Title: 17th DASC. AIAA/IEEE/SAE Digital Avionics Systems

Conference. Proceedings

Conference Date: 31 Oct.-7 Nov. 1998 Conference Location: Bellevue,

WA, USA

Language: English Document Type: Conference Paper (PA)

Treatment: Applications (A); Practical (P)

Abstract: Advanced transportation systems can be expected to increasingly

employ modern software development methodologies for the implementation of

their operational capabilities. This is demanded by the shift to open

architecture hardware and software solutions in order to meet stringent

low-cost objectives. The common object request broker architecture (CORBA)

is an important maturing standard offering interoperability among software

components and portability across hardware platforms. While CORBA can

provide needed interoperability and portability, the question arises how to

manage distributed CORBA clients and servers in a fault tolerant manner.

Future mission critical and distributed software deployments require

effective resource allocation. Critical resource allocation issues that

need to be addressed include component migration, scheduling, and load

balancing across dynamic computational resources. A management system that

addresses these issues in a uniform and cost-effective way is needed. This

paper describes an integrated approach to optimal resource allocation and

fault tolerance in a distributed computing environment. The approach

leverages a commodity-off-the-shelf (COTS) resource allocation system to

provide the infrastructure for the specific resource allocation and fault

tolerance enhancements required to host a reliable CORBA applications on

heterogeneous networked workstations. (19 Refs)

Descriptors: aerospace computing; distributed object management; fault

tolerant computing; open systems; resource allocation; software engineering

Identifiers: resource allocation; distributed computing environment;

software development methodologies; open architecture; common object

request broker architecture; interoperability; portability; distributed

clients; distributed servers; fault tolerant manner; mission critical

deployments; distributed software deployments; component migration;

scheduling; load balancing; commodity-off-the-shelf; heterogeneous

networked workstations

Class Codes: C7460 (Aerospace engineering computing); C6110J (

Object-oriented programming); C5470 (Performance evaluation and testing);

C6110B (Software engineering techniques)

6219122 INSPEC Abstract Number: B1999-05-6150-016

Title: Resource sharing in advance reservation agents

Author(s): Schelen, O.; Pink, S.

Author Affiliation: Comput. Sci. & Electr. Eng., Lulea Univ. of Technol.,

Sweden

Journal: Journal of High Speed Networks vol.7, no.3-4 p.213-28

Publisher: IOS Press,

Publication Date: 1998 Country of Publication: Netherlands

CODEN: JHSNEB ISSN: 0926-6801

SICI: 0926-6801(1998)7:3/4L.213:RSAR;1-U

Material Identity Number: P611-1999-003

U.S. Copyright Clearance Center Code: 0926-6801/98/$8.00

Language: English Document Type: Journal Paper (JP)

Treatment: Practical (P)

Abstract: We present an architecture where clients can make end-to-end

resource reservations through agents. For each domain in the network, there

is an agent responsible for immediate and advance admission control.

Reservations from different sources to the same destination domain are

aggregated as their paths merge toward the destination. We show that

network resources can be shared between immediate and advance reservations

without being pre-partitioned. Admission control for immediate reservations

uses information about resources to be allocated for advance reservations

in the near future. An important parameter in the admission control

algorithm is the so called lookahead time, i.e., the point at which we

actually start making resources available for approaching advance

reservations by rejecting immediate requests. In our model, preemption of

immediate reservations is made in cases where the admission control cannot

make resources available through rejection of immediate requests. The risk

of preemption can be varied by changing the lookahead time. We explore,

with simulations, the effects of providing advance reservations with this

model. The results show the cost in terms of resource utilization,

rejection probability and preemption probability. (21 Refs)

Descriptors: probability; quality of service; resource allocation;

telecommunication congestion control

Identifiers: advance reservation agents; client reservations; admission

control; reservation aggregation; resource sharing; resource allocation;

lookahead time; simulations; rejection probability; preemption probability

Class Codes: B6150 (Communication system theory); B0240Z (Other topics

in statistics)

6119175 INSPEC Abstract Number: B9902-6210L-031, C9902-6150N-027

Title: A dynamic quality of service middleware agent for mediating

application resource usage

Author(s): Brandt, S.; Nutt, G.; Berk, T.; Mankovich, J.

Author Affiliation: Colorado Univ., Boulder, CO, USA

Conference Title: Proceedings 19th IEEE Real-Time Systems Symposium (Cat.

No.98CB36279) p.307-17

Publisher: IEEE Comput. Soc, Los Alamitos, CA, USA

Publication Date: 1998 Country of Publication: USA xiii+493 pp.

ISBN: 0 8186 9212 X Material Identity Number: XX98-03381

U.S. Copyright Clearance Center Code: 0 8186 9212 X/98/$10.00

Conference Title: Proceedings Real-Time Systems Symposium

Conference Sponsor: IEEE Comput. Soc. Tech. Committee on Real-Time Syst.;

Tech. Univ. Madrid

Conference Date: 2-4 Dec. 1998 Conference Location: Madrid, Spain

Language: English Document Type: Conference Paper (PA)

Treatment: Practical (P)

Abstract: High bandwidth applications with time-dependent resource

requirements demand certain resource level assurances in order to operate

correctly. Quality of service resource management techniques are being

successfully developed that allow network systems to provide such

assurances. These solutions generally assume that the operating system at

either end of the network is capable of handling the throughput

requirements of the applications. However, real operating systems have to

manage many concurrent applications with varying resource requirements.

Without specialized support, the operating system cannot guarantee the

resources needed for any particular application. In support of these kinds

of applications we have developed a middleware agent called a dynamic QoS

manager (DQM) that mediates application resource usage so as to ensure that

applications get the resources they need in order to provide adequate

performance. The DQM employs a variety of algorithms to determine

application resource allocations. Using application QoS levels, it provides

for resource availability based algorithmic variation within applications

and varying application periods. It also allows for inaccurate application

resource usage estimates through a technique we have developed called

dynamic estimate refinement. The paper discusses new developments in the

design of the DQM and presents results showing DQM performance with both

real and synthetic applications. (26 Refs)

Descriptors: client-server systems; network operating systems; quality of

service; real-time systems; resource allocation

Identifiers: dynamic quality of service middleware agent; mediating

application resource usage; high bandwidth applications; time-dependent

resource requirements; resource level assurances; quality of service

resource management techniques; network systems; operating system;

throughput requirements; real operating systems; concurrent applications;

resource requirements; dynamic QoS manager; application resource usage; DQM

; application resource allocations; application QoS levels; resource

availability based algorithmic variation; varying application periods;

inaccurate application resource usage estimates; dynamic estimate

refinement

Class Codes: B6210L (Computer communications); C6150N (Distributed

systems software); C5620 (Computer networks and techniques); C0310 (EDP

management); C6150J (Operating systems)

6306239 INSPEC Abstract Number: B1999-09-6250F-017, C1999-09-7410F-023

Title: A communicating scheme for adaptive resource allocation in wireless

networks

Author(s): Al Agha, K.; Zeghlache, D.

Author Affiliation: Lab. PRiSM, Univ. de Versailles, France

Conference Title: ICUPC '98. IEEE 1998 International Conference on

Universal Personal Communications. Conference Proceedings (Cat.

No.98TH8384) Part vol.1 p.65-9 vol.1

Publisher: IEEE, New York, NY, USA

Publication Date: 1998 Country of Publication: USA 2 vol. xxxiii+1377

pp.

ISBN: 0 7803 5106 1 Material Identity Number: XX-1998-03128

U.S. Copyright Clearance Center Code: 0 7803 5106 1/98/$10.00

Conference Title: ICUPC '98. IEEE 1998 International Conference on

Universal Personal Communications. Conference Proceedings

Conference Date: 5-9 Oct. 1998 Conference Location: Florence, Italy

Language: English Document Type: Conference Paper (PA)

Treatment: Applications (A); Practical (P)

Abstract: To achieve the best possible bandwidth utilization in wireless

networks adaptive resource assignment is a requirement. Distributed or

centralized resource allocation schemes can be used to reach such an

objective. However, the means to actually implement these resource adaptive

features represent a major implementation challenge. In addition, the

schemes should be self adaptive and eventually self learning since large

scale traffic variations on an hourly and a daily basis exhibit patterns

with low expected variance. At a much finer scale, in micro and pico

cellular environments variations due to user mobility are unpredictable and

only dynamic channel allocation methods can adequately adapt to the rather

large traffic variations. This contribution relates to the larger scale

traffic variations and proposes an adaptive resource assignment approach

based on the multi-agent technique which relies on self learning. The

adaptive resource allocation scheme "channel segregation (CS)" containing

inherently the self learning attribute was selected to test and validate

the concept of using intelligent agents in base stations. Using the

multi-agent platform DIMA (development and implementation of multi agent

system), CS has been integrated and implemented in a simulation model of

base stations to introduce the self learning feature in the network. The

obtained simulation results validate the intelligent agent implementation

and illustrate the feasibility of using such agents in base stations to

introduce the self adaptive feature in mobile networks. (9 Refs)

Descriptors: adaptive systems; channel allocation; digital simulation;

microcellular radio; multi-agent systems; picocellular radio; radio

networks; self-adjusting systems; telecommunication computing; unsupervised

learning

Identifiers: communicating scheme; adaptive resource allocation; wireless

networks; bandwidth utilization; centralized resource allocation;

distributed resource allocation; self adaptive scheme; self learning scheme

; large scale traffic variations; variance; pico cellular environment;

micro cellular environment; user mobility; dynamic channel allocation;

multi-agent technique; channel segregation; intelligent agents; base

stations; multi-agent platform DIMA; development and implementation of

multi-agent; simulation model; mobile networks

Class Codes: B6250F (Mobile radio systems); C7410F (Communications

computing); C6185 (Simulation techniques); C6170K (Knowledge engineering

techniques)

6558121 INSPEC Abstract Number: C2000-05-7210N-047

Title: Generic agent framework for Internet information systems

Author(s): Erni, A.; Norrie, M.C.; Kobler, A.

Author Affiliation: Inst. for Inf. Syst., Swiss Fed. Inst. of Technol.,

Zurich, Switzerland

Conference Title: Information Systems in the WWW Environment. IFIP

TC8/WG8.1 Working Conference p.151-70

Editor(s): Rolland, C.; Chen, Y.; Fang, M.

Publisher: Kluwer Academic Publishers, Dordrecht, Netherlands

Publication Date: 1998 Country of Publication: Netherlands vi+376 pp.

ISBN: 0 412 82980 0 Material Identity Number: XX-1999-01751

Conference Title: Proceedings of Information Systems in the WWW

Environment

Conference Date: 15-17 July 1998 Conference Location: Beijing, China

Language: English Document Type: Conference Paper (PA)

Treatment: Practical (P)

Abstract: For effective Internet database services, it is essential that

the information requirements of regular users can be met without the

typical delays currently experienced using Internet browsers and the World

Wide Web. We use cooperating agents to manage both client and server

caches, thereby bringing significant performance improvements. The caching

and prefetching of information is based on both user and application

profiles and agents communicate to ensure the currency of client caches.

According to specific application requirements, various forms of agents can

be installed on the server and client sides to provide value-added services

to both casual and regular users. All component agents are instantiations

and/or specialisations of a generic agent. We describe how a specific

Internet brokering system for engineering product data has been constructed

using our general framework for the development. (17 Refs)

Descriptors: cache storage; client-server systems; cooperative systems;

information resources; information systems; Internet

Identifiers: generic agent framework; Internet information systems;

information requirements; delays; Internet browsers; World Wide Web;

cooperating agents; client server system; caches; performance improvements;

value-added services; Internet brokering system; engineering product data

Class Codes: C7210N (Information networks); C6170 (Expert systems and

other AI software and techniques); C1230 (Artificial intelligence); C6150N

(Distributed systems software); C6120 (File organisation)

5909996 INSPEC Abstract Number: C9806-6150J-002

Title: Transition-optimal token distribution

Author(s): Chlebus, B.S.; Diks, K.; Pelc, A.

Author Affiliation: Inst. of Inf., Warsaw Univ., Poland

Journal: Fundamenta Informaticae vol.32, no.3-4 p.313-28

Publisher: IOS Press,

Publication Date: Dec. 1997 Country of Publication: Netherlands

CODEN: FUINE8 ISSN: 0169-2968

SICI: 0169-2968(199712)32:3/4L.313:TOTD;1-9

Material Identity Number: J943-98003

Language: English Document Type: Journal Paper (JP)

Treatment: Practical (P)

Abstract: There is given a graph, that models a communication network of

a multiprocessor system, and there are tokens (jobs) allocated to nodes of

the graph. The task is to distribute the tokens evenly, subject to the

constraint that they may be moved only along the edges of the graph. The

cost of a distribution strategy is measured as the total number of

operations of moving a token along an edge. An algorithm for general graphs

is developed, by reduction to a maximum-flow minimum-cost problem, that

finds a cost-optimal distribution strategy, given a graph and an initial

token allocation. The main result is an algorithm for graphs that are lines

of nodes; it finds the distribution strategy in time O(n), for a line of n

nodes. (11 Refs)

Descriptors: multiprocessor interconnection networks; resource allocation

Identifiers: transition-optimal token distribution; communication network

; multiprocessor system; distribution strategy; maximum-flow minimum-cost

problem; cost-optimal distribution strategy

Class Codes: C6150J (Operating systems); C4230M (Multiprocessor

interconnection); C5440 (Multiprocessing systems)

5651180 INSPEC Abstract Number: C9709-6150N-060

Title: Contracting and moving agents in distributed applications based on

a service-oriented architecture

Author(s): Schulze, B.; Madeira, E.R.M.

Author Affiliation: Inst. of Comput., Univ. Estadual de Campinas, Sao

Paulo, Brazil

Conference Title: Mobile Agents. First International Workshop, MA '97

Proceedings p.74-85

Editor(s): Rothermel, K.; Popescu-Zeletin, R.

Publisher: Springer-Verlag, Berlin, Germany

Publication Date: 1997 Country of Publication: Germany viii+221 pp.

ISBN: 3 540 62803 7 Material Identity Number: XX97-00864

Conference Title: Mobile Agents. First International Workshop, MA '97.

Proceedings

Conference Date: 7-8 April 1997 Conference Location: Berlin, Germany

Language: English Document Type: Conference Paper (PA)

Treatment: Practical (P)

Abstract: The paper presents a service-oriented platform for the

development and execution of distributed applications based on contracting

stationary and migrating services. Services are seen as active objects

build on top of middleware using OMG/CORBA and added features. Customized

services add to the middleware the ability to handle transparently

application start-up and distribution according to load-balancing and

inverse caching application demand. Services can be considered of any kind

ranging from scientific specialized processing to data archiving

juke-boxes. An application on system management in scientific experimental

environment drives the work on some aspects of the architecture. (25 Refs)

Descriptors: cache storage; distributed processing; object-oriented

programming; resource allocation; software agents; software engineering

Identifiers: agent movement; agent contracting; distributed application

development; distributed application execution; service-oriented

architecture; stationary services; migrating services; active objects;

middleware; OMG/CORBA; customized services; application start-up;

application distribution; load balancing; inverse caching application

demand; system management; scientific experimental environment

Class Codes: C6150N (Distributed systems software); C6170 (Expert

systems); C6110J (Object-oriented programming); C6110B (Software

engineering techniques); C6120 (File organisation)

5672678 INSPEC Abstract Number: C9710-6170-015

Title: Multi-agent collaboration in competitive scenarios

Author(s): Fuchs, F.

Author Affiliation: Tech. Univ. Munchen, Germany

Conference Title: Re-Engineering for Sustainable Industrial Production.

Proceedings of the OE/IFIP/IEEE International Conference on Integrated and

Sustainable Industrial Production p.275-83

Editor(s): Camarinha-Matos, L.M.

Publisher: Chapman & Hall, London, UK

Publication Date: 1997 Country of Publication: UK viii+535 pp.

ISBN: 0 412 79950 2 Material Identity Number: XX97-01943

Conference Title: Re-Engineering for Sustainable Industrial Production.

Proceedings of the OE/IFIP/IEEE International Conference on Integrated and

Sustainable Industrial Production

Conference Date: May 1997 Conference Location: Lisbon, Portugal

Language: English Document Type: Conference Paper (PA)

Treatment: Applications (A); Practical (P)

Abstract: For many multi-agent scenarios one can assume that the agents

behave cooperatively and contribute to a common goal according to their

design. However, our work focuses on competitive scenarios which are

characterized by the agents' strong local interests their high degree of

autonomy, and the lack of global goals. Therefore, two agents will

cooperate if, and only if, both will gain-or at least expect to gain-from

that cooperation. This paper presents a conflict resolution mechanism which

is appropriate for competitive resource allocation in dynamic environments

which is based on compromising. It integrates a goal relaxation mechanism,

negotiation histories, and multilateral negotiations. (13 Refs)

Descriptors: artificial intelligence; cooperative systems; expert systems

; resource allocation

Identifiers: multi-agent collaboration; competitive scenarios;

multi-agent scenarios; goal relaxation mechanism; negotiation histories;

multilateral negotiations

Class Codes: C6170 (Expert systems); C6150J (Operating systems); C1230

(Artificial intelligence)

6011869 INSPEC Abstract Number: C9810-6170-009

Title: A reference-based cooperative algorithm for sharing distributed

resources

Author(s): Hisano, T.

Journal: Bulletin of the Electrotechnical Laboratory vol.61, no.10

p.33-41

Publisher: Agency Ind. Sci. & Technol. Minist. Int. Trade & Ind., Ibaraki

Publication Date: 1997 Country of Publication: Japan

CODEN: DESIA7 ISSN: 0366-9092

SICI: 0366-9092(1997)61:10L.33:RBCA;1-C

Material Identity Number: B061-98004

Language: English Document Type: Journal Paper (JP)

Treatment: Practical (P); Theoretical (T)

Abstract: We describe a reference-based cooperative algorithm for the

general problem in which every agent (semi-autonomous software module or

process) tries to occupy more than one distributed resource simultaneously.

The algorithm assumes no centralized control mechanism, but local data that

can be written by one agent and read by other agents. We show that the

agents executing the cooperative algorithm never fall into a deadlock or a

starvation. The interaction between non-cooperative agents and cooperative

agents executing the reference-based algorithm is also discussed in order

to find out the minimum function of agents in a truly distributed

environment. It is shown that joining one agent without the function of

mutual concession to cooperative agents may cause a starvation. (12 Refs)

Descriptors: concurrency control; cooperative systems; distributed

algorithms; resource allocation; software agents

Identifiers: reference-based cooperative algorithm; sharing distributed

resources; semi-autonomous software module; minimum function of agents;

truly distributed environment; starvation; deadlock; subproblems;

multiagent systems; cooperative computing; software paradigm; chatting

researchers problem; extended dining philosophers problem; faulty algorithm

; polite agent

Class Codes: C6170 (Expert systems); C6150J (Operating systems); C6150N

(Distributed systems software); C4240P (Parallel programming and algorithm

theory)

5513332 INSPEC Abstract Number: C9704-7210-011

Title: A simple computational market for network information services

Author(s): Mullen, T.; Wellman, M.P.

Author Affiliation: Artificial Intelligence Lab., Michigan Univ., Ann

Arbor, MI, USA

Conference Title: ICMAS-95 Proceedings. First International Conference on

Multi-Agent Systems p.283-9

Publisher: AAAI Press, Menlo Park, CA, USA

Publication Date: 1995 Country of Publication: USA xvii+472 pp.

ISBN: 0 262 62102 9 Material Identity Number: XX96-02140

Conference Title: Proceedings of First International Conference on

Multiagent Systems

Conference Date: 12-14 June 1995 Conference Location: San Francisco,

CA, USA

Language: English Document Type: Conference Paper (PA)

Treatment: Economic aspects (E); Theoretical (T)

Abstract: Visionary projections of a wide-area network teeming with

intelligent agents describe an environment where end-users and their agents

can pick and choose among a great variety of potentially valuable

information services. However, neither network capabilities nor users' time

and money are infinite. Computational markets provide one type of mechanism

for allocating limited resources in such an environment in a distributed,

dynamic way. Moreover, the underlying economic theory provides an

analytical framework for predicting aggregate behavior and designing

individual agents. We describe a prototypical computational market model

for information services distributed over a network. Our initial focus is

on the economic problem of when and where to establish mirror sites for the

more popular information services. Competitive agents choose to set up

mirrors based on going prices for network bandwidth, computational

resources, and the information service. Depending on the experimental

setup, we observed a range of qualitative behaviors. (21 Refs)

Descriptors: cooperative systems; information services; marketing;

resource allocation; software agents; wide area networks

Identifiers: computational market; network information services;

wide-area network; intelligent agents; resource allocation; economic theory

; aggregate behavior; mirror sites; competitive agents

Class Codes: C7210 (Information services and centres); C1290D (Systems

theory applications in economics and business); C6170 (Expert systems);

C1230 (Artificial intelligence)

4940251 INSPEC Abstract Number: C9506-6150N-045

Title: Comparative evaluation of cooperative plan execution strategies in

multiagent environments

Author(s): Lauzac, S.; Znati, T.F.

Author Affiliation: Dept. of Comput. Sci., Pittsburgh Univ., PA, USA

p.126-35

Publisher: IEEE Comput. Soc. Press, Los Alamitos, CA, USA

Publication Date: 1995 Country of Publication: USA xii+351 pp.

ISBN: 0 8186 7091 6

U.S. Copyright Clearance Center Code: 1080-241X/95/$4.00

Conference Title: Proceedings of Simulation Symposium

Conference Sponsor: SCS; IEEE Comput. Soc.; ACM

Conference Date: 9-13 April 1995 Conference Location: Phoenix, AZ, USA

Language: English Document Type: Conference Paper (PA)

Treatment: General, Review (G); Practical (P)

Abstract: Coordinating the action of distributed agents in a dynamically

changing environment is a difficult task. The difficulty stems from the

fact the real-world environment changes independently from the actions of

the agents. Furthermore, the agents may have limited competence and

awareness about one another's decisions and about the environment. This

paper focuses on cooperative plan execution and presents a dynamic

threshold-based scheme for balancing load among distributed agents.

Different load balancing strategies, namely client-initiated,

server-initiated, and hybrid, are discussed, and a comparative analysis of

these strategies is presented. The study shows that dynamic threshold-based

strategies perform well compared to static threshold-based methods. The

study also shows that dynamic threshold-based strategies greatly reduce the

communication overhead, and are well suited for cooperative plan execution

in dynamic environments. (11 Refs)

Descriptors: client-server systems; cooperative systems; planning

(artificial intelligence); resource allocation

Identifiers: cooperative plan execution strategies; multiagent

environments; coordinated action; distributed agents; dynamically changing

environment; dynamic threshold-based scheme; load balancing;

client-initiated strategy; server-initiated strategy; hybrid strategy;

comparative analysis; communication overhead

Class Codes: C6150N (Distributed systems software); C1230 (Artificial

intelligence); C6170K (Knowledge engineering techniques)