Abstract: A method for partitioning keys onto radix tree logical pages and a parallel index page build algorithm in order to provide radix tree build speedup proportional to the number of processors on the system and controlled efficient page utilization. Also, since keys are intelligently partitioned so that a complete set of keys is inserted into a logical page, there is no page overflow during the tree construction and thus page splitting is eliminated. Since radix index trees are really groups of logical pages in which each logical page contains a small tree, the tree is built (with respect to the logical pages) from the bottom up, while within each individual logical page the tree is constructed from the top down. The space required for a logical page is pre-allocated to allow construction of limbs to begin without waiting for the build of their underlying pages to complete.

Abstract: A browser for efficiently browsing large directory trees is presented. The browser uses authentication links as the structure through which the browser navigates. By adhering to the rules for a valid authentication chain, the browser increases efficiency by storing the results of preliminary steps to browsing.

Abstract: A method is provided for determining how semistructured information is organized in disparate semistructured resources by providing a wrapper to extract information and to provide structured information (e.g., tuples of an SQL database) to a mapper coupled to a standard relational database engine. In a specific embodiment, a querying agent is provided on top of the mapper. Further according to the invention, structured high-level user queries are processed across the disparate semistructured resources using a plurality of wrappers each dedicated to a particular resource.

Abstract: An apparatus for capturing and storing a co-marketer identification symbol representing an identity of an entity that has referred a user on a user station to a computer service, wherein the user station is coupled to the computer service by a communications path. A database is provided for storing a plurality of user records. Each of the user records includes a user identification field for storing information uniquely associating each of the user records with a user, and a co-marketer identification field for storing identity information representing the identity of an entity that directed the user to the computer service. An enrollment means is coupled to the communications path and the database, and is provided for enrolling a user on the computer service. The enrollment means includes means for determining a co-marketer that directed the user to the computer service, and means for assigning a unique user identification number to the user.

Abstract: A file system includes a file allocation table (FAT) and a file allocation table extension. Each FAT element of the file allocation table corresponds to a unique cluster on disk and represents the status of that cluster. These FAT elements are indexed by a cluster number that corresponds to that cluster on disk. Each element of the file allocation table extension does not correspond to any region on disk. Writing a file divided into clusters to disk first involves determining whether the cluster may be represented by a hole. If not, the cluster is written to disk at a particular cluster number and a status indicator is stored in the file allocation table at a FAT element indexed by that cluster number indicating that the cluster is allocated. If the cluster may be represented by a hole, a status indicator is stored in the file allocation table extension at an element indicating that the cluster is not stored on the disk of the computer.

Abstract: Method and apparatus for creating reliably duplicatable tape volumes by padding the volume label on the source volume with compression resistant random data, and copying such a source volume to a tape volume that does not have padding. In a preferred embodiment, the application program creating the source volume uses the present invention to create the padding on the source volume. The amount of padding to be inserted can be specified either as a fixed amount, or as a dynamically variable parameter changed on the basis of statistically significant results sampling. When a padded source volume is then duplicated to another tape, without padding, the likelihood that the source data will "fit" on the unpadded destination volume becomes much more predictable.

Abstract: An object-oriented programming language includes support for encapsulation of aspects and for change notification for aspect values. These features are then used to implement various change notification mechanisms. In general, a notification request for change in an aspect value in a target object is forwarded to any object (which may be the same object or another object) having aspects from which the aspect value is computed, derived, or replicated.

Abstract: Information "feeds" received from multiple information sources are aggregated, re-formatted and stored in a personalized real time information display system. When the display system receives a request from a user having a personal computer or other similar display device, the system is arranged to respond with information needed to display a screen saver that includes personalized, up to date information, such as traffic, weather and sports, that is of interest to that particular user. Each user has a personalized user profile indicating (a) the categories or types of information that the user desires to receive, such as sports information, weather, investment advisories, and so on, and (b) parameters that specify, for that user, the exact information desired to be received in each category.

Abstract: A computer system for creating a semantic object model from an existing relational database schema. The computer system analyzes the catalog information of the relational database schema and creates a semantic object for each table defined in the catalog. For each column defined within a table, a simple value attribute is added to the semantic object created for the table. The system then analyzes the relationship information stored in the catalog to create object link attributes that define relationships between two or more semantic object as well as to create multivalued group attributes and multivalued, simple value attributes. If the database catalog does not include the relational information, the user is prompted to indicate related semantic objects.

Abstract: Data is migrated from source media to target media in an automated tape library, where identifiers associated with the data are obtained from the source media, and the data and associated identifiers are consolidated by storing data/identifier paris in substantially continuous form or the target media. A plurality of source media units, such as magnetic tape cartridges, are first loaded to the library. Each source media unit contains one or more data blocks and a machine-readable identifier associated with each data block. The machine-readable identifier, for example, may be embodied by a machine-readable code imprinted on a housing of the media unit, or a code stored on the media itself. The identifier for each data block is obtained using the corresponding source media. Multiple data blocks and identifiers may be consolidated onto a single output media unit.

Abstract: A data system has a data server and multiple clients. The data server organizes data according to a tree index structure, where memory pages used to store data are indexed by higher level index nodes in the tree structure. The index nodes are replicated and maintained locally at the clients. The data organization on the server is further characterized by use of indexed side links between data pages to provide side access traversal, such as a Pi-tree structure. During a search for a particular search space, a requesting client traverses its own index replica until reference is made to a data page at the server. If the request causes a data page split or otherwise changes the storage location of a particular search space, the server sends information back as part of the result message to the requesting client to update the tree index replica. However, no coherence messages are sent to other clients.

Abstract: This invention is directed to relational extenders for a computer-based relational database. Each relational extender includes at least one column, in a first, business, table containing a user defined application database, dedicated to object handles for defining the complex data type of an object; a second, attribute, table containing at least one column defining a unique characteristic associated with the one object and one column dedicated to containing the object handle; and a third, metadata, table containing at least one column defining a common characteristic associated with all objects defined within the business table and one column dedicated to containing the object handle and at least one column dedicated to containing a reference to object data associated with the object.

Abstract: A method and apparatus for extending a relational database management system performed by a computer to manage object data stored in an object server having a processor and a data storage device. The method comprises the steps of receiving a database command including a store command and object data from a client at a federated coordinator, generating a data surrogate unique to the object data, associating the data surrogate with the object data, and storing the data surrogate and the object data in the data storage device in the object server.

Abstract: Systems and methods of time based data expiration in a replication environment are presented. The systems and methods of the present invention allow different age limits to be set on each individual server in a network. A network can therefore be configured with some servers having no age limit and other servers having a wide range of differing age limits. Systems and methods for handling mixed data sets that include some data objects which are allowed to expire and some data objects which do not expire are also presented. In all cases, the time based data expiration systems and methods are designed to insure proper operation of the replication process and any associated data recovery process. Specifically, expired data will not be recovered via any data recovery mechanism and servers will not attempt to recover missing data from a server that has expired that data.

Abstract: During operation of a computer system, index records and data records of a data structure are organized in a hierarchical manner, with index records being at a higher level of the hierarchy than the data records referenced by the index records. While the data structure is manipulated, the data records are stored in a random access memory. Modifications made to the data structure are chronologically written to a disk storage in a bottom-first/top-last order of the hierarchy for persistent storage as a database. A back-up copy of the database is made by storing the data and index records of the database in a top-first/bottom-last order on a sequentially readable media. A temporal and spatial view of a portion of the database can be obtained by accessing the index records and the data records of the database or the back-up media in the top-first/bottom last order.

Abstract: A method and apparatus for adding and deleting data in a plurality of heaps stored in memory, where the ordering of the data in the heaps is maintained across the heaps as the additions and deletions are performed. Not all entries in the heaps have a corresponding entry in all of the other heaps, but the ordering of the data in the heaps is in accordance with an order dependent list stored in memory. Such ordered heaps are useful for example, in a incremental linker software program.

Abstract: A system and method for identifying class relationships in an object oriented system. An object oriented program in an object oriented source code language is parsed for the immediate class relationship data. The immediate class relationship data is stored in a relation data library. In response to a user query, a derive engine is employed to derive multiple level class relationship data from the immediate class relationship data. The multiple class relationship data is presented on a system display to the user. As the number of immediate class relationships captured by the parser will be overwhelmingly huge in the case of a huge object oriented system, the derivation step is preferably divided into two. In anticipation of the user query, the immediate relationship data for each class of object is preprocessed and stored in a facts library according to class and type of relationship.

Abstract: A method for record searching in a database within a computer system are disclosed. The computer system includes a main processor, a main memory, and a peripheral storage device having a secondary processor. In accordance with the method of the present invention, a command block specifying a search string for record searching in at least one database table of the database is prepared. Subsequently, the command block is issued from the main processor to the secondary processor within the peripheral storage device of the computer system. The secondary processor within the peripheral storage device is then utilized to read the database table(s) into a memory within the peripheral storage device, in response to a receipt of the command block. The search string in the command block is compared to each record of the database table(s) within the memory of the peripheral storage device to identify all the records therein which contain the search string.

Abstract: A method and system for reclaiming stacked volumes within a peripheral data storage subsystem is disclosed. In accordance with the method and system of the present invention, a database is interrogated to determine whether or not an opportunistic reclaim threshold of a stacked volume has been reached after a service request from a host computer has been performed on a stacked volume and while the stacked volume is still mounted. If the opportunistic reclaim threshold of the stacked volume has not been reached, the stacked volume is released and dismounted. However, if the reclaim threshold of the stacked volume has been reached, another determination is made as to whether or not a service request for the host computer is pending. If the host computer requires service, the stacked volume is again released and dismounted. Otherwise, if the host computer does not require service, at least one data set from the mounted stacked volume is opportunistically reclaimed while the stacked volume is still mounted.

Abstract: A method and system for automatically synchronizing a target and source directory in a computer system are provided. The system (8) includes a first file directory (32) which has a first plurality of files (33). A second file directory (70) has a second plurality of files (71). A directory synchronizer (14) sequentially selects one of the first plurality of files (33), determines whether one of the second plurality of files (71) is substantially similar to the selected one of the first plurality of files (33), and copies the selected one of the first plurality of files (33) into the second file directory (70) if no similar file is found.