Andrew Tokmakoff

Discovery and composition of services for context-aware systems

We consider the challenge of dynamically adapting services to context changes that occur in ubiquitous computing environments (e.g., changes in a users activity) and propose the Context-Aware Service Enabling (CASE) platform for that purpose. The CASE platform combines context-aware service discovery with service composition, acting as an enabler for the development of adaptive context-aware applications. In this paper, we illustrate the need for context-aware service discovery and composition in pervasive 4G environments and present the architecture of the CASE platform. The CASE platform enables applications to easily adapt to changes in service availability, which may result from changes in client and/or service context. We also provide an overview of the platforms technical realization.

Security architecture for open collaborative environment

The paper presents proposed Security Architecture for Open Collaborative Environment (OCE) being developed in the framework of the Collaboratory.nl (CNL) project with the intent to build a flexible, customer-driven security infrastructure for open collaborative applications. The architecture is based on extended use of emerging Web Services and Grid security technologies combined with concepts from the generic Authentication Authorization and Accounting (AAA) and Role-based Access Control (RBAC) frameworks. The paper describes another proposed solution the Job-centric security model that uses a Job description as a semantic document created on the basis of the signed order (or business agreement) to provide a job-specific context for invocation of the basic OCE security services. Typical OCE use case of policy based access control is discussed in details.

Job-centric security model for open collaborative environment

This paper describes the design and development of a flexible, customer driven, security infrastructure for open collaborative environments. The experiences were gained within the framework of the collaboratory.nl project. The work is based on extended use of emerging Web services and grid security technologies, combined with concepts from the generic authentication authorization and accounting (AAA) authorisation framework. Basic CNL use cases and functional security requirements are analysed to provide motivation for the proposed job-centric security model. This model describes access control and user- and resource management. The proposed job-centric approach uses a job description as a semantic document that is created on the basis of the signed order (or business agreement). It contains all the information required to run the experiment and also to create/manage the virtual job-based associations of users and resources. The proposed trust relations analysis explains the use of trust anchors in the job-centric security model. In addition, the paper provides implementation details of using XACML and SAML for authorisation assertions and messaging, based on the current CNL implementation

An Approach to the Analysis of Interworking Traders

The ODP Trader provides a match-making service for Objects in an Open Distributed System. In previous work, a model of the Trader was created using Coloured Petri Nets, incorporating its capacity for standalone and co-operative Interworking amongst multiple Trader instances. In this paper, the CPN model of the Trader is analysed using Equivalence Classes (for Occurrence Graph reduction) when the Trader is configured as a standalone entity servicing multiple requests concurrently. The Trader is also analysed in a number of interworking scenarios which are, in turn, used for reasoning about complex Trader interworking topologies.

Service trading in mobile environments

One of the most important considerations for applications operating in a mobile environment is the ability to dynamically adjust to changing resource availability. With a dynamic resource location service as part of the infrastructure, mobile applications are able to re-bind to service providers as they move within the mobile environment and as available services change. As part of the ISO/IEC and ITU-T standardisation of the Reference Model for Open Distributed Processing (RM-ODP), a trading function has been identified. It is an important infrastructure service that allows clients to locate services using service types and associated parameters as selection criteria. In order to engineer reliable open object-based distributed systems, it is important to perform modelling and analysis as part of the design process. This paper introduces the reader to the concept of service trading and indicates how it may be utilised by applications in a mobile environment. In addition, coloured Petri nets are shown to be useful for the modelling and verification of distributed mobile applications.

Coloured Petri net modelling of the ODP Trader for use in resource discovery

As high speed networking becomes more widespread and increasingly accessible from home, a huge variety of new vendors and services will become available to consumers. These services will fall into three main categories: information services, shopping and entertainment. One of the major problems facing an increasingly information-rich society is how to quickly and conveniently locate information which is provided by different servers throughout a large distributed system. The Reference Model for Open Distributed Processing (RM-ODP) deals with the problem of locating services/resources in a heterogeneous open distributed system through the standardization of a Trading function which is realized by a Trader. To increase our understanding of a Traders dynamic behaviour, a CPN model has been created. The Traders computational viewpoint has been modelled using a hierarchy of CPNs which allow functional blocks to be modelled separately and their inter-relationships defined. A discussion of the model and techniques for its analysis are presented.

Designing a material analysis collaboratory for the process industry

Cost and efficiency are important incentives for academic oriented collaboratories, where researchers and analysts can work and experiment with advanced remote equipment. Such collaboratories are not suitable for industrial use, because they include experimental instruments instead of of-the-shelf instruments, and because industry requires other security and reliability implementation. This paper presents an industrial collaboratory that aims to support the daily work of material analysts. It supports natural collaboration between analysts and with customers and supports remote control of sophisticated of-the-shelf instruments. The paper includes a sketch of the technical architecture and elaborates on business models for commercial exploitation of the collaboratory