Distributed Computation as an Economic System

Abstract

A s everyone knows, computer networks play an increasingly important role in the economy: in the daily operation of big firms, the smooth running of government agencies, and transactions in financial markets. However, it may surprise even professional economists to find that economics is of help in allocating computational resources to programs that run in these systems.The reason is that computer networks can be regarded as a community of concurrent processes, or a "computational ecosystem" (Huberman and Hogg, 1988). As these networks grow, they become analogous to human market economies because the processes-in their interactions, strategies, and lack of perfect knowledge-face the same issues as people in a market. This paper will explain this analogy and explore one of its main implications: that economics may offer new ways to design and understand the behavior of these emerging computational systems. In particular, markets provide examples of methods to deal successfully with coordinating asynchronous operations in the face of imperfect knowledge. A computational system set up along market rules can allow the system as a whole to adapt to changes in the environment or disturbances to individual members. In what follows we will describe how market-based computational systems are already working and how they may sometimes break down. Some of the best-known programs already using these ideas include computational resource allocation (Malone et al., 1988;Waldspurger et al., 1992;Nash, 1993) and thermal markets for controlling building environments (Clearwater and Huberman, 1993), which have shown increased performance when compared with traditional operating systems.

Decentralized computation procurement and computational robustness in a smart market

“…Huberman and Hogg (1995) and Huberman, Lukose, and Hogg (1997) have developed the idea of computational portfolios, where a number of dierent techniques are simultaneously tried for solving the same problem. The idea is that on a multiprocessing computer that can support a number of simultaneous programs, choosing a level of time sharing between a number of solution techniques is similar to choosing a portfolio of stocks in ®nancial theory.…”

Markets as Global Scheduling Mechanisms: The Current State

“…2 The advent of networks of computers did not alienate scheduling from economics, and scheduling for distributed computer systems has also been recognized as an activity under resource management [2,12]. Markets for computing resources have been embraced by many as the key components in the operation of distributed systems [1,3,4,5,10,11,14,15,18,19,20,21,22,23]. …”

“…A computer network consists not only of multiple users, but also of multiple service providers, appearing much more complex [19] and closer to a market as we know from everyday life [1,5,10,11,15,21]. The seeming ease with which markets allocate resources, albeit its complexity [1,5,11,15,19], and the solid existence of theoretical microeconomics (particularly, the general equilibrium theory) [5,11,15,19], have captured the imagination of researchers in the field of networks.…”

“…The seeming ease with which markets allocate resources, albeit its complexity [1,5,11,15,19], and the solid existence of theoretical microeconomics (particularly, the general equilibrium theory) [5,11,15,19], have captured the imagination of researchers in the field of networks. We examine below the representative market models for distributed computing systems (the Contract Net Protocol [18], the Enterprise [14], a model by Kurose and Simha [11], Agoric System [15], Spawn [10,21], WALRAS [3,22,23], Mariposa [19,20], and the Grid Architecture for Computational Economy [1]), 3 focusing on the ones that make an extensive and direct use of the general equilibrium theory: the model by Kurose and Simha, and WALRAS.…”

Approximating the 0–1 Multiple Knapsack Problem with Agent Decomposition and Market Negotiation

“…Market-oriented algorithms have been applied successfully towards solving complex resource allocation problems such as distributed multi-commodity flow in the trucking industry (Wellman 1993(Wellman , 1996, query processing and data migration in a distributed database (Stonebraker et al 1994), task scheduling in a distributed operating system (Huberman 1995, Malone et al 1988, Miller and Drexler 1988, network and file system resource allocation (Ferguson et al 1988, Gagliano and Mitchem 1996, Kurose and Simha 1989, Kuwabara et al 1996, Miller et al 1996, and Yemini 1981, allocating tradeable pollution permits (Marron and Bartels 1996), task allocation in discrete manufacturing systems (Baker 1996, Tilley 1996, and load balancing in distributed systems (Ferguson et al 1996). Most of the current research has focused on developing one of two types of economic models.…”