Abstract: A method for accessing a shared resource is provided. An assigned usage rate is received from a resource coordinator and a desired usage rate is determined. When it is determined that the desired usage rate is higher than the assigned usage rate, a shared resource may be accessed at an enhanced usage rate if a usage reserve has been accumulated. When a shared resource is accessed at an enhanced usage rate, the usage reserve is decremented by an amount based on a difference between the enhanced usage rate and the assigned rate. When there is no usage reserve accumulated, access to the shared resource is limited to the assigned usage rate. When the desired usage rate is not higher than the assigned usage rate, a shared resource is accessed at the desired usage rate. When the desired usage rate is less then the assigned usage rate, the usage reserve is accumulated up to a reserve maximum. The reserve maximum may be based on configuration data.

Abstract: A method and apparatus are provided for performing incremental refreshes to materialized views defined by one-to-N lossless joins. Each base table of the materialized view is selected to be processed as the current "selected table". The processing of the current selected table varies depending on whether the selected table is the right side table of an outer join. If the selected table is not the right table of an outer join, then the selected table is processed by (1) deleting rows from the materialized view based on rows of the selected table that have been updated or deleted in the selected table during the batch window, and (2) inserting rows into the materialized view based on updates and inserts into the selected table that occurred during the batch window. If the selected table is the right table of an outer join, then changes made to the selected table are processed in a way that reduces the number of changes that have to be made to the materialized view.

Abstract: Every node in a multi-node parallel database system maintains a logical clock for generating timestamps. The logical clocks are synchronized by attaching a current timestamp to every message that is sent by a node. When a node receives an incoming timestamp that is greater than the value indicated by the associated logical clock, it sets the associated logical clock forward to at least the value of the timestamp. Each node continually sends and receives a message including a timestamp at least once in a prescribed period of time. Moreover, each node maintains a high watermark that represents a time greater or equal to the highest time of any logical clock in the multi-node parallel database system.

Abstract: A method and apparatus that allow schema version evolution to occur without requiring applications that expect older schema versions to be recompiled is provided. According to one aspect of the invention, each application that requests data is supplied the data in the format that the application expects. To supply the data in the expected format, a mechanism is provided for tracking the evolution of data types without losing information. In addition, mechanisms are provided for determining the format expected by the application and the format in which the data is currently stored. A mechanism is also provided for converting the data from the stored format to the expected format when the two formats do not match. A data migration strategy is described in which data is gradually migrated to newer formats when the data is updated by applications that expect a more recent format than the format in which the data is currently stored.

Abstract: A method and system for delivering a live feed to a client is provided. According to one aspect of the invention, content data is generated by an encoder. Tag data that indicates locations of video frame data within the content data is generated while the content data is being generated. According to one embodiment, the tag data is generated by the encoder. According to an alternative embodiment, the tag data is generated by parsing the content data. The content data is at a location from which the content data is delivered to the client. The tag data is stored at a location from which the tag data may be used to provide the client non-sequential access to the content data. Before the encoder finishes generating the content data, a request is received for non-sequential access to the content data by the client, second content data is constructed based on the content data, the tag data and the request for non-sequential access, and the second content data is sent to the client.

Abstract: A multithreaded client application initiates multiple database operations associated with at least one transaction. The client application allocates and stores a separate runtime context for each of the database transactions. These transaction runtime contexts are also distinct from the state information, or runtime context, of the thread initiating a particular operation of the transaction. As a result, operations started by a thread that terminates may be continued, or resumed, by a subsequent thread without having to reperform previously executed operations. The duration of the transactions' runtime context is thus decoupled from the duration of the thread's runtime context. Special syntax and routines are provided for explicitly allocating a runtime context and deallocating a runtime context.

Abstract: One embodiment of the present invention provides a method for constructing a query for an object-relational database. This method displays at least part of a schema for the object-relational database to a user in a graphical representation. Next, the method allows the user to perform graphical manipulations on the schema. The method then generates a query based upon the schema and the graphical manipulations, and this query is sent to the object-relational database. Another embodiment of the present invention provides a method for processing an object-relational query for transmission from a client to an object-relational database server. This method receives the query at the client, and parses the query to determine which object-relational structures the query is attempting to access.

Abstract: A virtual bookshelf system provides a user a way to browse and locate information associated with a plurality of documents. The virtual bookshelf system contains classification information to classify the documents in a hierarchical structure that includes a plurality of hierarchical levels based on pre-defined categories. The virtual bookshelf system displays a portion of the hierarchical structure including a representation of the categories. A user of the virtual bookshelf system may browse through the categories in the hierarchical levels to learn the general topics of the available documents. The user is permitted to select a particular category, and in response, the virtual bookshelf system displays document information for documents classified in the category selected. When selected, peripheral themes or issues are displayed such that the themes relate to the main issues covered by the documents classified under the selected category.

Abstract: An apparatus and process for storing information objects in a database are disclosed. An information object, instantiated according to an abstract data type having a plurality of attributes, is pickled into an image that contains all values stored in the object in a linearized representation. In one embodiment, the entire image is stored in a single column of a table of the database. In another embodiment, each portion of the image associated with one of the attributes is stored in a separate column of the database table.

Abstract: A method and apparatus for I/O forwarding in a cache coherent shared disk computer system is provided. According to the method, a requesting node transmits a request for requested data to a managing node. The managing node receives the read request from the requesting node and grants a lock on the requested data. The managing node then forwards data that identifies the requested data to a disk controller. The disk controller receives the data that identifies the requested data from the managing node and reads a data item, based on the data that identifies the requested data, from a shared disk. After reading the data item from the shared disk, the disk controller transmits the data item to the requesting node. In one embodiment, an I/O destination handle is generated that identifies a read request and a buffer cache address to which the data item should be copied.

Abstract: An apparatus and process for storing representations of information objects having null values in a database are disclosed. A null object is established in memory associated with an information-bearing object. The null object has an attribute null value corresponding to each data-bearing attribute of the information-bearing object. Each attribute null value is set to a value indicating whether the corresponding attribute of the information-bearing object is null, not null, or not capable of having a null value. The null object also has an atomic null value that indicates whether the entire information-bearing object is null. When the information-bearing object is to be accessed, the null object is first accessed to determine whether the entire information-bearing object or any attributes thereof are null. Accordingly, unnecessary retrieval and manipulation of the information-bearing object is avoided.

Abstract: A method and apparatus for is provided for efficiently transmitting multiple rows of data to a client station. The client station generates a query that is transmitted to a remotely located server, which in turn generates a query result. The server transmits the first row of data from the query result to the client. The remaining rows of data are transmitted by first calculating the difference between successive rows of data. Next, the server sends only the unique rows resulting from the calculated row difference to the client. The server also send an indication of the number of non-repeated value columns detected in order to allow reconstruction of the query result by the client station.

Abstract: A method and system for providing non-sequential access to visual information that is being digitally encoded in a digital data stream is provided. The digital data stream includes a sequence of video frame data. Each video frame data in the sequence of video frame data corresponds to a video frame of the visual information. The digital data stream is generated with an encoder. The encoder generates tag data that indicates locations of the video frame data within the digital data stream. The digital data stream is stored at a location from which the digital data stream is delivered to a client. Tag data is stored at a location from which the tag data may be used to provide the client non-sequential access to the digital data stream. A selected set of video frames within the digital data stream is selected based on the tag data in response to a request for non-sequential access by the client.

Abstract: Methods and systems are provided for providing locking in a system. The resource objects of the system can be persistent, and thus provide finer granularity locking by allowing shared resource objects to be dynamically allocated and de-allocated. The persistent nature of the resource objects allows for the dynamic allocation and de-allocation of the resource objects, while guarding against the abnormal release of resource locks that result in the possibility of inconsistent data. If abnormal release occurs, the affected resource object(s) are marked to indicate that they are in a "dubious" state. The dubious resource objects are not dynamically de-allocated, but are instead allowed to persist until corrective or validation actions are taken. If the locks are validly released, then the resource objects can be safely de-allocated.

Abstract: A record management system is provided for generating a multi-dimensional view for different measures. A set of records is retrieved in response to a set of queries. The records include dimension values and measure values, which are associated with the measures. The set of records is maintained in a master table. The record management system generates a record structure foundation, including a query map and a master table index. The record management system then employs the record structure foundation to generate a multi-dimensional layout mapping for the measures that are to be represented in the view. The record structure foundation and measure values in the master table are used to convert the layout mapping into the desired multi-dimensional view. The record management system includes data storage units for implementing the master table, query map, master table index, and layout mapping.

Abstract: A method and apparatus for passing runtime contexts between threads in a multithreading application is provided. A precompiler is provided which allows programmers to declare and use runtime contexts in the same manner as other conventional variable types are declared and used. Special syntax and routines are provided for explicitly allocating a runtime context and deallocating a runtime context, thus decoupling the duration of a runtime context from the duration of the thread that initiated the operations whose states are stored in the runtime context. Consequently, operations started by a thread that terminates may be continued by a subsequent thread without having to re-perform previously executed operations. Precompiler directives are provided to explicitly state which runtime contexts will be used for certain operations. A single thread can therefore use one or more runtime contexts.

Abstract: A method and apparatus for presenting and modifying data from a set of tables in a database is provided. A view that is defined is based on a set of one or more tables that may include relational tables or object tables. The view defines a presentation of data from the one or more tables as a set of objects that reside in the database. Data is read from the one or more rows of the tables based on the view, and is presented as a set of objects that reside in the database. An object id that is based on data from the one or more rows is generated and associated with each object presented. The view may specify which columns from the one or more tables contain values used to generate the object ids. A trigger may associated with the view. The set of objects presented may be presented as objects having an attribute that is a column object. Column objects include user specified object types, collection objects (e.g. nested tables and variable arrays), or references to objects.

Abstract: A consistent set of data is supplied to a software application from databases. When a particular set of data is identified, a first process is requested to obtain a snapshot time from a database server associated with the first database. The snapshot time causes all subsequent reads of the first database by the first process to return data that reflects a database state associated with the snapshot time. A first set of data in the first database is locked in order to produce a copy of data from a first database. After locking the first set of data, a plurality of processes are requested to obtain snapshot times from a database server associated with the first database. The snapshot times cause all subsequent reads of the first database by the plurality of processes to return data from the first database as of the snapshot times.

Abstract: An efficient set of indexes to cover a plurality of anticipated query types is determined by building a directed acyclic graph whose nodes correspond to anticipated query types. A minimum leaf spanning tree for the equivalent graph is determined by repeatedly finding an augmenting path for a current spanning tree and producing a reduced leaf spanning tree based on the current spanning tree and the augmenting path until an augmenting path can no longer be found. The leaves of the minimum leaf spanning tree indicate which indexes should be built.

Abstract: A method and apparatus for multi-phase locking for partition maintenance operations. In the first phase, a shared data dictionary lock is acquired on a body of metadata in a data dictionary. Next, the data dictionary is read and a list of affected partitions is generated. In the second phase, an intent exclusive data lock is acquired on the affected table. Next, an exclusive data lock is acquired on the affected partitions of the affected table. The shared data dictionary lock on the data dictionary is released and a physical attribute of the data of the affected partitions is changed. In the third phase, an exclusive data dictionary lock is acquired on the data dictionary. The metadata associated with the affected partitions in the data dictionary is updated and the exclusive data locks on the affected partitions and the intent exclusive data lock on the affected table are released. Finally, the exclusive data dictionary lock on the data dictionary is released.