Towards increased expressiveness in service level agreements: Research Articles

  • Authors:
  • Viktor Yarmolenko;Rizos Sakellariou

  • Affiliations:
  • School of Computer Science, The University of Manchester, Kilburn Building, Oxford Road, Manchester M13 9PL, U.K.;School of Computer Science, The University of Manchester, Kilburn Building, Oxford Road, Manchester M13 9PL, U.K.

  • Venue:
  • Concurrency and Computation: Practice & Experience - Middleware for Grid Computing: A “Possible Future”
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

The aim of this paper is to argue for the benefits of increasedexpressiveness in service level agreements (SLAs). Such benefitsmay be obtained from the use of analytical expressions to specify aSLA's agreement terms as functions and not as variable or constantvalues or ranges. The main idea behind this thinking is thatfunctions may contain variables defined in the SLA or be drawn fromthe known set of reference variables, such as wall-clock time, jobstart time, current bandwidth of the resource, etc. Experiences andconclusions drawn are in the context of SLA-based job managementsystems. We demonstrate that the use of analytical expressions inSLAs can potentially reduce the overheads associated with jobrenegotiation and/or reduce the number of failed agreements.Copyright © 2006 John Wiley & Sons, Ltd.