SPAWN - A DISTRIBUTED COMPUTATIONAL ECONOMY

被引：251

作者：

WALDSPURGER, CA ^{[1
]}

HOGG, T ^{[1
]}

HUBERMAN, BA ^{[1
]}

KEPHART, JO ^{[1
]}

STORNETTA, WS ^{[1
]}

机构：

[1] XEROX CORP, PALO ALTO RES CTR, DYNAM COMPUTAT GRP, PALO ALTO, CA 94304 USA

来源：

IEEE TRANSACTIONS ON SOFTWARE ENGINEERING | 1992年 / 18卷 / 02期

关键词：

CONCURRENT SYSTEMS; DISTRIBUTED SYSTEMS; DYNAMIC LOAD SHARING; MICROECONOMIC ALGORITHMS; PRIORITY MECHANISMS; RESOURCE ALLOCATION; SCHEDULING;

D O I：

10.1109/32.121753

中图分类号：

TP31 [计算机软件];

学科分类号：

081202 ; 0835 ;

摘要：

We have designed and implemented an open, market-based computational system called Spawn. The Spawn system utilizes idle computational resources in a distributed network of heterogeneous computer workstations. It supports both coarse-grain concurrent applications and the remote execution of many independent tasks. Using concurrent Monte Carlo simulations as prototypical applications, we explore issues of fairness in resource distribution, currency as a form of priority, price equilibria, the dynamics of transients, and scaling to large systems. In addition to serving the practical goal of harnessing idle processor time in a computer network, Spawn has proven to be a valuable experimental workbench for studying computational markets and their dynamics.

引用

页码：103 / 117

页数：15

共 38 条

[1]

Agha GA, 1990, ACTORS MODEL CONCURR

[2] A DISTRIBUTED LOAD-BALANCING POLICY FOR A MULTICOMPUTER [J].

BARAK, A ;

SHILOH, A .

SOFTWARE-PRACTICE & EXPERIENCE, 1985, 15 (09) :901-913

[3]

CHOW TCK, 1982, IEEE T SOFTWARE ENG, V8, P401

[4]

Coffman Jr E. G., 1973, OPERATING SYSTEMS TH

[5] NEGOTIATION AS A METAPHOR FOR DISTRIBUTED PROBLEM-SOLVING [J].

DAVIS, R ;

SMITH, RG .

ARTIFICIAL INTELLIGENCE, 1983, 20 (01) :63-109

[6]

DUPUY A, 1986, NETWORK SIMULATION T

[7]

ELLIS JR, 1985, YALEUDCSRR364 YAL U

[8]

ENGELBRECHTWIGG.R, 1983, AUCTIONS BIDDING CON

[9]

FERGUSON D, 1988, P 8 INT C DISTR COMP, P491

[10]

FRIEDMAN D, 1984, AM ECON REV, V74, P60

← 1 2 3 4 →