Abstract: A computer system for performing searches on a collection of information includes a mechanism through which results from a search query are ranked according to user specified relevance factors to allow the user to control how the search results are presented. The relevance factors are applied to the results achieved for each query. That is, each item returned by the search has a set of attributes. Each of these attributes is assigned a weight according to the specified relevance factors. These weights are combined to provide a score for the item. Search results are provided to the user, ordered according to scores. The application of the relevance factors does not alter the query performed on the collection of information.

Abstract: Disclosed is a server architecture for connecting to a plurality of remote client computers each seeking access to applications resident on the server. The architecture provides (1) a reliable, single point of entry for client users that enforces security and recovers immediately from failures; (2) connection management that monitors the user's level of online activity, terminating inactive connections both to save system resources and to limit unnecessary connection charges to the user; (3) application management that spreads users optimally among active application instances, maintaining a pool of available applications, initiating new instances when the pool is low, and which records a user's utilization of different applications for billing purposes; (4) system integrity, which is achieved through division of transaction-management responsibility among the various interrelated components of the invention; and (5) continuous, uninterrupted operation.

Abstract: Disclosed is an object-oriented approach to storage and transmission of retrievable items in a client-server computer environment. Special processing modules manage retrieval and permanent storage, on client computers, of requested objects in a manner that minimizes unnecessary transmission and storage. The user enters a "subscription" in connection with items related to hierarchically inferior items. The subscription specifies the amount of information the user desires to download and store locally with respect to a specified hierarchical depth level. The invention also includes a pruning module for deleting, from local storage, items inconsistent with the subscription and which the user does not explicitly indicate a desire to store, and a facility for updating locally stored items upon re-establishment of communication with the server.

Abstract: Design of indexes in a relational database management system is based on a workload analysis of all requests in a system. Each request is assigned a value of importance, and requests are broken into expressions, contexts and columns to facilitate identification of candidate indexes. Candidate indexes are compared to existing indexes to determine whether the existing indexes can be reused or modified. Candidate indexes are classified as hashed or sorted. Based on the index design, record placement is selected to be hashed, sorted or by default. Related indexes and tables are clustered and the clusters are sized to fit file areas. Cache buffers are also defined.