Design and implementation of economics-based resource management system in ad hoc grid

https://doi.org/10.1016/j.advengsoft.2011.10.003Get rights and content

Abstract

The ad hoc grid is a spontaneous organization of cooperating heterogeneous nodes into a logical community without a fixed infrastructure and with only minimal administrative requirements. Resource management for ad hoc grids is challenging due to the participation of heterogeneous, dynamic, autonomous and ephemeral grid nodes. The paper proposes an ad hoc grid resource management system, the producers and consumers of ad hoc grid resource are modeled as the self-interested decision-makers described in microeconomic theory. All market participants in the ad hoc grid environment including grid resources and services can be represented as agents. We apply economic agents to build ad hoc grid resource management, where ad hoc grid resource consumers and providers can buy and sell ad hoc grid resource based on an underlying economic architecture. The main processes involved in ad hoc grid resource management are resource registration, discovery, and resource allocation. The experiments are conducted to compare ad hoc grid resource allocation algorithm with other ad hoc grid resource allocation algorithm. Simulation results show that our proposed algorithm is more efficient than compared allocation scheme.

Introduction

The ad hoc grid is a spontaneous organization of cooperating heterogeneous nodes into a logical community without a fixed infrastructure and with only minimal administrative requirements [1]. The main goal of an ad hoc grid is to provide computing resources on demand to every participant. Unlike traditional grid systems, the number of nondedicated grid nodes is much higher, demanding non-intrusive operation of the ad hoc grid middleware. Ad hoc grids allow a group of individuals to accomplish a mission that involves computation and communication among the grid components, often without fixed structure. Examples of applications of ad hoc grids include: disaster management, wild fire fighting, and defense operations. An ad hoc grid allows grid entities, also referred to as ad hoc grid peers, to spontaneously establish an ad hoc relationship, join existing grids, dynamically contribute services to the grid, and invoke services offered by other peers in the grid. Ad hoc grids facilitate interaction in an autonomous fashion without requiring pre-configured environments or management policies. They support a large class of applications that cannot be conventionally supported by traditional grid environments. These applications include market-oriented applications, transient collaborations, sporadic interactions, and other community applications that require on-the-fly grid establishment and deployment. In an ad hoc grid, every node in the network can spontaneously arise as a resource consumer or a resource producer at any time when it needs a resource or it possesses an idle resource. Ad hoc grids are highly heterogeneous and dynamic, one of the main challenges of resource allocation in such environments is to find mechanisms which do not rely on the global information and are robust to the changes in resource availability in grid. Ad hoc grids are characterized by heterogeneity, autonomy, and volatility [2]. These characteristics result in varying workload of the resource manager in the ad hoc grid. Therefore it is required to develop a resource allocation mechanism that can balance the workload of the resource, and can enable the ad hoc grid to self-organize itself.

Resource management for ad hoc grids is challenging due to the participation of heterogeneous, dynamic, autonomous and ephemeral grid nodes. Due to limited resources (energy, bandwidth, etc.), user preferences may lead to an unfair load distribution in ad hoc grids. The situation worsens, in cases where ad hoc grid nodes act both as grid clients and grid resources from time to time but refuse to contribute resources. The paper proposes an ad hoc grid resource management system, the producers and consumers of ad hoc grid resource are modeled as the self-interested decision-makers described in microeconomic theory. All market participants in the ad hoc grid environment including grid resources and services can be represented as agents. We apply economic agents to build ad hoc grid resource management, where ad hoc grid resource consumers and providers can buy and sell ad hoc grid resource based on an underlying economic architecture. The agents of the ad hoc grid system interact by offering to buy or sell resources at given prices. The main processes involved in ad hoc grid resource management are resource registration, discovery, and resource allocation. The experiments are conducted to compare ad hoc grid resource allocation algorithm with other ad hoc grid resource allocation algorithm.

The rest of the paper is structured as followings. Section 2 discusses the related works. Section 3 gives the overview of system model. Section 4 presents economics based ad hoc grid resource management. Section 5 describes ad hoc grid resource allocation algorithm. In section 6 the experiments are conducted and discussed. Section 7 gives the conclusions to the paper.

Section snippets

Related works

Currently, ad hoc grid is hot research topic. Shivle et al. [1] studied assigning resources in an ad hoc grid to an application composed of communicating subtasks. Their goal of the allocation is to minimize the average percentage of energy consumed by the application to execute across the machines in the ad hoc grid, while meeting an application execution time constraint. Six different heuristic approaches of varying time complexities have been designed and compared via simulations to solve

Model description

The overall ad hoc grid resource management system model consists of three layers (Fig. 1). The lower layer is the underlying ad hoc grid resource. Ad hoc grid resources on this layer are owned and allocated by ad hoc grid resource agents deployed at the nodes in the ad hoc grid. The top layer is the system’s interface to ad hoc grid user. The middle layer is the agent-based ad hoc grid resource management system. It consists of two types of agent and market institution that allocates resources

Economics based ad hoc grid resource management

The main actions involved in ad hoc grid resource management are ad hoc grid resource registration, ad hoc grid resource discovery, and ad hoc grid resource allocation. Ad hoc grid resource registration is activated when ad hoc grid resource agents register their presence by publishing their resource descriptions. The ad hoc grid resource discovery provides a mechanism to discover grid resources for grid task agents. Ad hoc grid resource allocation is used for grid task agents to obtain and

Ad hoc grid resource allocation algorithm

Ad hoc grid resource allocation is used for ad hoc grid task agent to access the located ad hoc grid resource. Once ad hoc grid task agent receives the reply description of ad hoc grid manager, it can start communication with the ADGM to use the services. Negotiation is the process by which ad hoc grid task agent and ad hoc grid resource agent interacts to reach an agreement through ad hoc grid market. In general, the negotiation consists of the following steps:

  • (1)

    Ad hoc grid task agents (ADGTA)

Experiments environment

In this section, we evaluate the performance of ad hoc grid resource allocation algorithm (ADGrid_RA) using the JAVASIM [23] simulator. Network generator BRITE [22] generates the computer network topology. We simulate ad hoc grid environment with a 2 dimension area of 500 m * 500 m to study mobile device’s behavior. Each mobile device in the simulated environment has a maximal radio range of 100 m, and moves following a random-walking mobility model. The average speed of each mobile device is 5 m/s.

Conclusions

The paper proposes an ad hoc grid resource management system. The agent-based ad hoc grid resource management system consists of two types of agent and market institution that allocates resources in response to the selling of ad hoc grid resource agent and buying behavior of ad hoc grid task agents. We apply economic agents to build ad hoc grid resource management, where ad hoc grid resource consumers and providers can buy and sell ad hoc grid resource based on an underlying economic

Acknowledgements

The work was partly supported by the National Natural Science Foundation of China (NSF) under grant (Nos. 60970064, No. 61171075), National Key Basic Research Program of China (973 Program) under Grant No. 2011CB302601, Open Fund of the State Key Laboratory of Software Development Environment under Grant (No. SKLSDE-2011KF-01), Fok Ying Tong Education Foundation, China (Grant No. 121067), Program for New Century Excellent Talents in University, China (NCET-08-0806). Any opinions, findings, and

References (23)

  • Abdullah Tariq, Sokolov Vassiliy, Pourebrahimi Behnaz, Bertels Koen. Self-organizing dynamic ad hoc grids. In: 2008...
  • Cited by (8)

    • Self-adaptive systems: A survey of current approaches, research challenges and applications

      2013, Expert Systems with Applications
      Citation Excerpt :

      Aggressive and conservative bidding strategies reinforce adaptation in the variations of resource availability in ad hoc grids. Li and Li (2012) proposed an ad hoc grid resource management system, the producers and consumers of the ad hoc grid resource are modeled on the self-interested decision-makers described in microeconomic theory. The AESOP (Karnousko et al., 2010) initiative comprises a SOA approach for monitoring and control of Process Control applications (batch and continuous process).

    • A study of research trends and issues in wireless ad hoc networks

      2015, Mobile Computing and Wireless Networks: Concepts, Methodologies, Tools, and Applications
    • A study of research trends and issues in wireless ad hoc networks

      2014, Handbook of Research on Progressive Trends in Wireless Communications and Networking
    View all citing articles on Scopus
    View full text