Abstract: In parallel database management systems, database update requests typically result in activity at several nodes. Rollback of all updates is required if the update of any node fails. A coordination process monitors for failure or success of updates. The coordinator further provides for distinguishing activities that have taken place at any given node from the other nodes for different database update requests. Savepoints are local. This allows rollback of a selected update without affecting nodes which did not process the update.

Abstract: The invention establishes the context in which data exchanged between dissimilar relational database management systems can be mutually understood and preserved, and data conversions can be minimized. The invention accomplishes this by establishing layers of descriptive information which isolate machine characteristics, levels of support software, and user data descriptions. Optimized processing is achieved by processing the different descriptor levels at different times during the development and execution of the database management systems. Minimal descriptive information is exchanged between the cooperating database management systems. Any data conversions that may be necessary are done only by the receiver of the data, and only at the point where it is necessary to have the data represented in the receiver's native format for processing.

Abstract: The present invention provides an evaluation strategy for the execution of SQL queries involving recursion and table queues in a relational database management system. The table queue is established for the query in the memory of the computer, and tuples are generated by tuple-producing operators in the query for storage into the table queue. As operators of the query read the stored tuples from the table queue, a first end-of-file condition, termed a Temporary End-Of-File, occurs when all of the tuples currently stored in the table queue have been retrieved and control is returned to the operator reading from the table queue. A retry request for additional tuples is generated by the operator through the table queue to the tuple-producing operators. If the tuple-producing operators can generate additional tuples, then these additional tuples are stored into the table queue and control is returned to the operator reading from the table queue in response to the retry request.

Abstract: The invention establishes the context in which data exchanged between dissimilar relational database management systems can be mutually understood and preserved, and data conversions can be minimized. The invention accomplishes this by establishing layers of descriptive information which isolate machine characteristics, levels of support software, and user data descriptions. Optimized processing is achieved by processing the different descriptor levels at different times during the development and execution of the database management systems. Minimal descriptive information is exchanged between the cooperating database management systems. Any data conversions that may be necessary are done only by the receiver of the data, and only at the point where it is necessary to have the data represented in the receiver's native format for processing.

Abstract: A pre-compiler for a host programming language contains a language dependent portion and a language independent portion. The language dependent portion extracts source file statements written in a supplemental language and converts them into a language independent format. The language independent statements are communicated to the language independent portion, which translates them into lists of tasks to be performed. These tasks are communicated back to the language dependent portion in a language independent format. The language dependent portion then translates the tasks into procedure calls in the host language.

Abstract: An interfacing system and method for use in computerized systems for intercommunication between application and database systems. A plurality of generic application program interfaces are employed as an interface for use in precompiler development as a runtime interface target to initialize data structures with data required to perform runtime database query language-related functions. The application program is isolated from the database kernel interface facilitating alteration of the database kernel interface without affecting the application program or user-developed precompilers. Parameters used by the interface are employed by applications written to a plurality of host languages. The applications indirectly use these data structures required by the kernel. The interface system supports serialized multiple thread access to the database kernel facilitating performance/functional benefits of multi-threaded applications.

Abstract: In a relational database system, a method is utilized that increases the performance of the sort operation. An optimizer routine in the relational database manager analyzes a user's complete query to determine whether the final sorted results can be used directly by the user as sorting occurs. If the sort results can be used, the sort results are sent to Relational Data Services in the relational database manager for output to the user. Depending upon the determination made by the optimizer routine, one of two output modes for the final sorted sequence string of data are selected by the relational database manager. In disk output mode, the last pass of the final sorted sequence spring is written to disk. In fast direct output mode, the records of data are sent to the user as the final sorted sequence string is being merged during the last pass. The system and method of this invention reduces the total sort time by eliminating the overhead of writing the disk during the fast direct output mode.