|
|
Computational Science, Engineering & Technology Series
ISSN 1759-3158
CSETS: 27
TRENDS IN PARALLEL, DISTRIBUTED, GRID AND CLOUD COMPUTING FOR ENGINEERING
Edited by: P. Iványi, B.H.V. Topping
Chapter 16 Negotiations in an Agent-based Grid Resource Brokering System K. Wasielewska 1, M. Ganzha 1,2, M. Paprzycki 1,3, M. Drozdowicz 1, D. Petcu 4, C. Badica 5, N. Attaoui 6, I. Lirkov 7 and R. Olejnik 81Systems Research Institute Polish Academy of Sciences, Warsaw, Poland
2University of Gdansk, Poland
3Warsaw Management Academy, Poland
4Western University of Timisoara, Romania
5University of Craiova, Romania
6University of Tetouan, Morocco
7Bulgarian Academy of Sciences, Sofia, Bulgaria
8CNRS, Lille, France Full Bibliographic Reference for this chapter
K. Wasielewska, M. Ganzha, M. Paprzycki, M. Drozdowicz, D. Petcu, C. Badica, N. Attaoui, I. Lirkov, R. Olejnik, "Negotiations in an Agent-based Grid Resource Brokering System", in P. Iványi, B.H.V. Topping, (Editors), "Trends in Parallel, Distributed, Grid and Cloud Computing for Engineering", Saxe-Coburg Publications, Stirlingshire, UK, Chapter 16, pp 355-374, 2011. doi:10.4203/csets.27.16
Keywords: software agents, grid, resource brokering and management, ontology, negotiations, service level agreement.
Summary
When considering integration of business into the grid infrastructure, users (physical persons, or institutions) can be divided into two groups, according to the role that they play: (i) owners (service providers) who make their resources accessible, and (ii) clients who search for resources to commission execution of a job. Note that these roles are interchangeable, as providers of one type of services may become consumers of
another. Separately, it should be pointed out that "business" computations are
usually conducted in a predefined order, and within a fixed time frame (contrary
to volunteer computing projects, e.g. the folding@home, where neither of these two
factors plays an important role). In order to assure availability of resources,
and timeliness of job execution, it is necessary to formalize the contract in
the form of a service level agreement ( SLA), followed by job
monitoring, as well as assessing the quality of service. Therefore, grid systems
that aspire to provide business solutions should also consider the SLA
management as part of their functionality. Here, the SLA is defined as
a contract between a user of a service and a service provider, specifying the
scope of the service and co-operation conditions offered during the period of
contract validity, to ensure the quality of service ( QoS). The SLA can be considered as a means to provide more flexibility, scalability and optimizing resource utilization, and, consequently, increase the reliability of the grid.
In our Agents in Grid (AiG) project, software agents are to provide a meta-level grid middleware, in which (a) agents work in teams (managed by LMaster agents), (b) all meta-information is ontologically demarcated and semantically processed, and (c) an economic model is to be based on autonomic SLA negotiations, and QoS monitoring.
The aim of this chapter is to present an overview of issues concerning autonomous (possibly agent-based) \textit{Service Level Agreement} negotiations. To achieve this goal, first five European research projects are discussed that represent different approaches to the problem of integration of business and grid architectures, as well as various methods of establishing the SLA between grid users. Second, the experience gained from these projects is used as a foundation of conceptualization of the SLA negotiations in the AiG project.
purchase the full-text of this chapter (price £20)
go to the previous chapter
go to the next chapter
return to the table of contents
return to the book description
purchase this book (price £95 +P&P) |
