supported by the EC IST Programme
CoreGRID European Research Network on Foundations, Software Infrastructures and Applications
for large scale distributed, GRID and Peer-to-Peer Technologies
Home arrow CoreGRID WG arrow Research Topics arrow Service Level Agreement  

spacer spacer
Main Menu
Integration Activities
Training & Education
CoreGRID & Industry
Mobility Portal
Trust&Security Portal
Collaboration Gateway
Other Collaborations
spacer spacer
spacer spacer
Service Level Agreement

Topic leaders

Omer Rana, Cardiff University, UK,
Wolfgang Ziegler, Fraunhofer Institute, Germany,

Short description

Many existing computational science (e-Science) projects make use of pre-defined resources (and services) -- often defined statically at the start of application execution. The use of a dynamic service registry has been recognized as being important, but few applicationusers are currently employing one directly. As such eScience projects mature, and need to utilize resources (and services) which are outside their direct control, it is necessary to make use of SLAs to define: (i) requirements that such an application would place on resources (and services) owned by another third party; (ii) check whether these requirements have been met during use. SLAs therefore provide the interface between application users making demands on resources, and application providers determining what should be made available for external use.

The complexity of an SLA can therefore vary from a static description of resource names (specified in terms of IP addresses) to complex contraints defined as functions that can be evaluated at deployment time. In the first instance, it would be useful to define and deploy something simple on existing Grid infrastructure -- as many applications have identified this as a requirement. For instance, some multimedia application require the need for co-allocation -- something that can be defined as a requirement for a multi-party SLA.

In the initial period the theme will be devoted to better understanding requirements of eScience projects that could benefit from the use of SLAs. This initial study will be based on a questionnaire sent to various projects to assess their potential uptake of an SLA-based resource (and service) use. Identifying what should be specified in this SLA, and the type of SLA they would need to be in place prior to application execution (for instance, do they need a multi-site SLA, can parameters in the SLA be defined in an application independent manner, etc).

The aims of this theme are to investigate how SLAs can be specified, managed, monitored, used and defined in eScience applications. The theme will be driven by a number of research challenges:
  • Research Challenge 1: What should be part of an SLA? How should this be encoded (rules, constraints, keywords)?
  • Research Challenge 2: What can we do with an SLA once it has been defined? -- e.g. capacity planning, pricing services, etc. This issue becomes especially relevant in the context of emerging infrastructure - such as Cloud computing - where virtualized resources are being made available to external users.
  • Research Challenge 3: What types of SLAs are most beneficial to an end user community in computational science (e-Science)? Can we encourage their use to facilitate greater resource sharing?
  • Research Challenge 4: How can business or institutional policy be mapped into operational SLAs?


Service Level Objectives, Quality of Service, Service Pricing and Electronic Markets, Capacity Planning, Electronic Contracts, Software Licensing, Business Models, Revenue Streams.


[Apr. 09] Grid 2009 will host the next Service Level Agreements in Grids Workshop. Submission dead-line July 26.

Last Updated ( Wednesday, 06 May 2009 )
spacer spacer
spacer spacer
© 2012 CoreGRID Network of Excellence - European Grid Research
Joomla! is Free Software released under the GNU/GPL License.
spacer spacer