Abstract: A system and method for failure recovery in a multiple processing node system are described herein. Each node can be adapted to store a backup copy of its database portion and/or results to disk storage or memory of at least one other node. In the event of a failure of a node, the replacement node can be adapted to transfer or copy the backup copy of the database portion of the failed node from the failed node's neighbors to the replacement node's disk storage or memory in between database operations. Before the transfer or copy of the backup copy is completed, the replacement node can be adapted to perform database operations in part on the portion of the backup copy the replacement node has already received and in part on the backup copy stored at the neighboring node(s).

Abstract: A system and methods for parallel processing of queries to one or more databases are described herein. One or more databases may be distributed among a subset of slave nodes of a global-results processing matrix. A query to the database may be generated using a query-based high-level programming language. The query-based source code then may be converted to intermediary source code in a common programming language and then compiled into a dynamic link library (DLL) or other type of executable. The DLL is then distributed among the slave nodes of the processing matrix, whereupon the slave nodes execute related portions of the DLL substantially in parallel to generate initial query results. The initial query results may then be provided to a master node of the global-results processing matrix for additional processing, whereby the master node is adapted to execute one or more associated portions of the DLL on the initial query results.

Abstract: A system and method for configuring a plurality of processing nodes into a parallel-processing database system are described herein. Each of a plurality of processing nodes connected by a network receive software and one or more configuration files related to the intended function of the processing node. The software may include homogeneous agent software, one or more library dynamic-link libraries (DLL), and the like. The configuration file is used to configure the homogeneous agent to operate as the intended node in a global-results processing matrix, a general-purpose query processing matrix, or a index-base query processing matrix. Another node or nodes may be configured to convert query-based programming code to intermediary source code in a common programming language and then compile the intermediary source code into a dynamic link library (DLL) or other type of executable.

Abstract: A system and method for configuring a plurality of processing nodes into a parallel-processing database system are described herein. Each of a plurality of processing nodes connected by a network receive software and one or more configuration files related to the intended function of the processing node. The software may include homogeneous agent software, one or more library dynamic-link libraries (DLL), and the like. The configuration file is used to configure the homogeneous agent to operate as the intended node in a global-results processing matrix, a general-purpose query processing matrix, or a index-base query processing matrix. Another node or nodes may be configured to convert query-based programming code to intermediary source code in a common programming language and then compile the intermediary source code into a dynamic link library (DLL) or other type of executable.

Abstract: A method for distributing and sorting data among a plurality of nodes is described herein. After receiving a portion of a data set (e.g., a database), each node sorts its portion and estimates a partitioning of the sorted dataset among the nodes based in part on its own sorted data portion. Each node then provides a representation of its estimated partition to a master node. The master node, using the provided estimated partitions, determines a tentative partitioning and submits the tentative partitioning to each node. Each node then determines the effect the tentative partitioning using its data portion. If the effect is acceptable for each node, the tentative partitioning plan is used to partition the data. Otherwise, the tentative partitioning plan is repeatedly revised by the master node and considered by the nodes having data portions until an acceptable or optimum partitioning is determined.

Abstract: A system and method for scheduling database operations to one or more databases in a parallel-processing database system are described herein. After a query server generates a dynamic-link library (DLL) or other executable representative of one or more database operations to a database, the query server notifies a scheduling services module of the generation of the DLL and submits the DLL to a query agent. The query agent notifies the scheduling services module of its receipt of the DLL. Based on any of a variety of considerations, the scheduling services module schedules a time of execution for the DLL by one or more processing matrices that store the database. At the scheduled time, the scheduling services module directs the query agent to submit the DLL to the indicated processing matrices. The scheduling services module also can be adapted to monitor the execution of previously submitted DLLs by one or more processing matrices and adjust the scheduled times of execution for subsequent DLLs accordingly.

Abstract: A system and method for configuring a plurality of processing nodes into a parallel-processing database system are described herein. Each of a plurality of processing nodes connected by a network receive software and one or more configuration files related to the intended function of the processing node. The software may include homogeneous agent software, one or more library dynamic-link libraries (DLL), and the like. The configuration file is used to configure the homogeneous agent to operate as the intended node in a global-results processing matrix, a general-purpose query processing matrix, or a index-base query processing matrix. Another node or nodes may be configured to convert query-based programming code to intermediary source code in a common programming language and then compile the intermediary source code into a dynamic link library (DLL) or other type of executable.

Abstract: A system and method for scheduling database operations to one or more databases in a parallel-processing database system are described herein. After a query server generates a dynamic-link library (DLL) or other executable representative of one or more database operations to a database, the query server notifies a scheduling services module of the generation of the DLL and submits the DLL to a query agent. The query agent notifies the scheduling services module of its receipt of the DLL. Based on any of a variety of considerations, the scheduling services module schedules a time of execution for the DLL by one or more processing matrices that store the database. At the scheduled time, the scheduling services module directs the query agent to submit the DLL to the indicated processing matrices. The scheduling services module also can be adapted to monitor the execution of previously submitted DLLs by one or more processing matrices and adjust the scheduled times of execution for subsequent DLLs accordingly.

Abstract: A system and methods for parallel processing of queries to one or more databases are described herein. One or more databases may be distributed among a subset of slave nodes of a global-results processing matrix. A query to the database may be generated using a query-based high-level programming language. The query-based source code then may be converted to intermediary source code in a common programming language and then compiled into a dynamic link library (DLL) or other type of executable. The DLL is then distributed among the slave nodes of the processing matrix, whereupon the slave nodes execute related portions of the DLL substantially in parallel to generate initial query results. The initial query results may then be provided to master node of the global-results processing matrix for additional processing, whereby the master node is adapted to execute one or more associated portions of the DLL on the initial query results.

Abstract: A system and methods for parallel processing of queries to one or more databases are described herein. One or more databases may be distributed among a subset of slave nodes of a global-results processing matrix. A query to the database may be generated using a query-based high-level programming language. The query-based source code then may be converted to intermediary source code in a common programming language and then compiled into a dynamic link library (DLL) or other type of executable. The DLL is then distributed among the slave nodes of the processing matrix, whereupon the slave nodes execute related portions of the DLL substantially in parallel to generate initial query results. The initial query results may then be provided to master node of the global-results processing matrix for additional processing, whereby the master node is adapted to execute one or more associated portions of the DLL on the initial query results.

Abstract: A system and method for failure recovery in a multiple processing node system are described herein. Each node can be adapted to store a backup copy of its database portion and/or results to disk storage or memory of at least one other node. In the event of a failure of a node, the replacement node can be adapted to transfer or copy the backup copy of the database portion of the failed node from the failed node's neighbors to the replacement node's disk storage or memory in between database operations. Before the transfer or copy of the backup copy is completed, the replacement node can be adapted to perform database operations in part on the portion of the backup copy the replacement node has already received and in part on the backup copy stored at the neighboring node(s).

Abstract: A system and method for configuring a plurality of processing nodes into a parallel-processing database system are described herein. Each of a plurality of processing nodes connected by a network receive software and one or more configuration files related to the intended function of the processing node. The software may include homogeneous agent software, one or more library dynamic-link libraries (DLL), and the like. The configuration file is used to configure the homogeneous agent to operate as the intended node in a global-results processing matrix, a general-purpose query processing matrix, or a index-base query processing matrix. Another node or nodes may be configured to convert query-based programming code to intermediary source code in a common programming language and then compile the intermediary source code into a dynamic link library (DLL) or other type of executable.

Abstract: A system and method for scheduling database operations to one or more databases in a parallel-processing database system are described herein. After a query server generates a dynamic-link library (DLL) or other executable representative of one or more database operations to a database, the query server notifies a scheduling services module of the generation of the DLL and submits the DLL to a query agent. The query agent notifies the scheduling services module of its receipt of the DLL. Based on any of a variety of considerations, the scheduling services module schedules a time of execution for the DLL by one or more processing matrices that store the database. At the scheduled time, the scheduling services module directs the query agent to submit the DLL to the indicated processing matrices. The scheduling services module also can be adapted to monitor the execution of previously submitted DLLs by one or more processing matrices and adjust the scheduled times of execution for subsequent DLLs accordingly.

Abstract: A system and methods for parallel processing of queries to one or more databases are described herein. One or more databases may be distributed among a subset of slave nodes of a global-results processing matrix. A query to the database may be generated using a query-based high-level programming language. The query-based source code then may be converted to intermediary source code in a common programming language and then compiled into a dynamic link library (DLL) or other type of executable. The DLL is then distributed among the slave nodes of the processing matrix, whereupon the slave nodes execute related portions of the DLL substantially in parallel to generate initial query results. The initial query results may then be provided to master node of the global-results processing matrix for additional processing, whereby the master node is adapted to execute one or more associated portions of the DLL on the initial query results.

Abstract: A method for distributing and sorting data among a plurality of nodes is describe herein. After receiving a portion of a data set (e.g., a database), each node sorts its portion and estimates a partitioning of the sorted dataset among the nodes based in part on its own sorted data portion. Each node then provides a representation of its estimated partition to a master node. The master node, using the provided estimated partitions, determines a tentative partitioning and submits the tentative partitioning to each node. Each node then determines the effect the tentative partitioning using its data portion. If the effect is acceptable for each node, the tentative partitioning plan is used to partition the data. Otherwise, the tentative partitioning plan is repeatedly revised by the master node and considered by the nodes having data portions until an acceptable or optimum partitioning is determined.