Abstract: A distributed system is described in which resource utilization decisions depend upon the categorization of resource descriptions stored in the distributed system. In the principal embodiment, the resource descriptions are web service descriptions which are augmented with tags (i.e. descriptive words or phrases) entered by users and/or by web service administrators. The system stores, for different groups of users, groups of tags popularly used by users within those groups. By monitoring tags input by a user, and proposing tags to the user from any groups which are under-represented in tags input by the user, a more balanced set of tags describing resources in the system is obtained. This leads to a more coherent and focussed set of tags in the system, which in turns results in better resource utilization decisions and hence a more efficient use of the resources of the distributed system.

Abstract: The present invention relates to distributed systems in which resource utilisation decisions depend upon the semi-automatic categorisation of resource descriptions stored in the distributed system. In the principal embodiment, the resource descriptions are web service descriptions which are augmented with tags (i.e. descriptive words or phrases) entered by users and/or by web service administrators. The initial use of automatic categorisation of these descriptions, followed by a user-driven fine-tuning of the automatically-generated categories enables the rapid creation of reliable categorisation of the resource descriptions, which in turns results in better resource utilisation decisions and hence a more efficient use of the resources of the distributed system.

Abstract: A distributed system is described in which resource utilisation decisions depend upon the categorisation of resource descriptions stored in the distributed system. In the principal embodiment, the resource descriptions are web service descriptions which are augmented with tags (i.e. descriptive words or phrases) entered by users and/or by web service administrators. A service taxonomy is constructed on the basis of the tags, and is subsequently used in suggesting tags to the user which the user might want to ascribe to the service in question. This leads to a more coherent and focused set of tags in the system, which in turns results in better resource utilisation decisions and hence a more efficient use of the resources of the distributed system.

Abstract: A peer-to-peer network operating in accordance with a service-oriented architecture is disclosed. The peers in the network request services from one another and each keep a record of the quality of service they receive from the other peers. The peers can operate in two modes of service provider selection. In a first mode, the selection is so as to favor service providers which have provided the peer with good service in the past. In a second mode, the selection is probabilistic and can therefore select service providers other than those that have provided good service in the past. Each device keeps track of the relative success of adopting the second mode of selection. By occasionally using the second mode of selection and adopting the second mode of selection more frequently should the relative success of adopting the second mode of selection rise, a more rapid adjustment by the network to changing network conditions is enabled.

Abstract: A peer-to-peer network operating in accordance with a service-oriented architecture is disclosed. The peers in the network request services from one another and each keeps a record of the quality of service they receive from the other peers. The peers share quality of service information with one another in order to take advantage of the improvement in the overall efficiency of the use of resources in the network offered by such information sharing. However, the invention provides a further improvement in that peers do not report the quality of service offered by the peers they have received the best quality of service from. This is found to increase the overall level of service still further since it prevents the peers converging on a favorite service provider and thereby overloading it. The invention finds particular application in distributed applications which dynamically select a Web Service to perform a function at run-time.

Abstract: A distributed system is described in which resource utilisation decisions depend upon the categorisation of resource descriptions stored in the distributed system. In the principal embodiment, the resource descriptions are web service descriptions which are augmented with tags (i.e. descriptive words or phrases) entered by users and/or by web service administrators. The system stores, for different groups of users, groups of tags popularly used by users within those groups. By monitoring tags input by a user, and proposing tags to the user from any groups which are under-represented in tags input by the user, a more balanced set of tags describing resources in the system is obtained.

Abstract: A distributed system is described in which resource utilisation decisions depend upon the categorisation of resource descriptions stored in the distributed system. In the principal embodiment, the resource descriptions are web service descriptions which are augmented with tags (i.e. descriptive words or phrases) entered by users and/or by web service administrators. A service taxonomy is constructed on the basis of the tags, and is subsequently used in suggesting tags to the user which the user might want to ascribe to the service in question. This leads to a more coherent and focussed set of tags in the system, which in turns results in better resource utilisation decisions and hence a more efficient use of the resources of the distributed system.

Abstract: The present invention relates to distributed systems in which resource utilisation decisions depend upon the semi-automatic categorisation of resource descriptions stored in the distributed system. In the principal embodiment, the resource descriptions are web service descriptions which are augmented with tags (i.e. descriptive words or phrases) entered by users and/or by web service administrators. The initial use of automatic categorisation of these descriptions, followed by a user-driven fine-tuning of the automatically-generated categories enables the rapid creation of reliable categorisation of the resource descriptions, which in turns results in better resource utilisation decisions and hence a more efficient use of the resources of the distributed system.

Abstract: A peer-to-peer network operating in accordance with a service-oriented architecture is disclosed. The peers in the network request services from one another and each keep a record of the quality of service they receive from the other peers. The peers can operate in two modes of service provider selection. In a first mode, the selection is so as to favour service providers which have provided the peer with good service in the past. In a second mode, the selection is probabilistic and can therefore select service providers other than those that have provided good service in the past. Each device keeps track of the relative success of adopting the second mode of selection. By occasionally using the second mode of selection and adopting the second mode of selection more frequently should the relative success of adopting the second mode of selection rise, a more rapid adjustment by the network to changing network conditions is enabled.