FULLTITLE
Service Negotiation

KEYWORDS
Negotiation, SOA, QoS

SUMMARY
The ongoing adoption and commercialisation of web and grid Service Oriented Architectures (SOA) has increased the pressure to develop new high level service support. With web and grid service architectures converging it is becoming clear that it is the higher level functionality that now needs addressing.
Whilst rudimentary mechanisms exist for dynamic service discovery in the form of technologies like UDDI, it is clear that service negotiation and monitoring have been relatively neglected. It is also clear that services implementations cannot hope to gain popularity if negotiation for use of a given resource has to occur “out of bounds”. Given service negotiation, and agreement on a given contract/SLA it is equally important that some way of monitoring the adherence to a given contract / SLA is possible.

Previous attempts to translate heavily worded legal documents into a workable electronic format capable of adaptation to specific circumstances have met with limited success. The ALDUS project (1990) ALDUS was enacted to examine possible areas of automation specifically with mind to the creation of sales contracts. At the conclusion of the project, however, it was decided that there were no viable economically feasible products upon which such tools could be built at the time. Fortunately this situation is changing and a number of new projects are starting to address the problem in depth.

Service negotiations in the current marketplace generalise to two different situations:

• Free Services – Often seen in scientific institutions; common during the development of new tools and services. This model is suitable only in business situations where client membership remains within a single administrative and organisational boundary. Services do not necessarily have to provide Quality of Service (QoS) guarantees, leading to a best effort situation in most circumstances. Given the trend towards distributed computing through VO development this model is untenable for widespread use.

• Economically supported services – A more feasible business model but brings with it additional concerns. An economic model requires attention to the following infrastructure considerations:
• Service discovery
     - The discovery and active differentiation of services
• Service negotiation
     - The requirement specification and capability specification respectively. The determination of a compromise situation between the two parties.
• Service Agreement
     - The signing of documents to guarantee service level attributes. This could involve a contract / SLA
• Service mediation
     - The specification of a standardised complaint and renegotiation policy
• Service monitoring
     - The monitoring of service use, based on data provided by a combination of client, service and possibly trusted third parties.

Traditionally service contracts are encountered in two forms, mainly dependent on the size and cost of a given contract.

• Standardised contracts based on a service classification. For example, Gold, Silver, Bronze etc. In these situations no customisation occurs, the possibilities for service requirement-capability matching is severely limited.

• Manual service negotiation, through meetings between client, service and legal aid. This model is most applicable to the VO vision as the negotiation can dynamically determine the most accurate compromise position possible. However, such processes clash with the need to maintain agility in the business process. It is this type of service negotiation that TRANSACT aims to emulate.

The DIRC grid project addressed these issues through the creation of the TRANSACT (Tool for Real-time Negotiation of Secure Authorisation Contracts) architecture. Over three years a prototype implementation making use of a number of different forms of requirement input was created. The output of this research can be seen in the papers listed below.

LINKS

PAPERS

Lock R. Contracts in a Grid Age.
Conference Proceedings 8th Cabernet Radicals Workshop. Ajjacio, Corsica. 5-8th October 2003

Lock R. Automated Contract Negotiation for the Grid.
Conference Proceedings, Prep 2004. University of Hertfordshire. 5-7th April

Lock R. TRANSACT (Tool for Real-time Automation of Secure Authorisation Contracts).
Dependability interdisciplinary research Collaboration (Internal Annual Project Conference) Conference Proceedings,
Newcastle, 29th-31st March 2004.

Lock R. Contract Negotiation, a Tale of Five Themes.
Dependability interdisciplinary research Collaboration (Internal Annual Project Conference) Conference Proceedings,
Nesc (National e-Science centre) Edinburgh 15-17th March 2005. pp66-67.
ISBN: 1-86220-159-5

Lock R. Automated negotiation for service contracts.
Conference Proceedings, to appear Compsac2006 Chicago 17th-21st September 2006

Author

Russell Lock (r dot lock at comp dot lancs dot ac dot uk)