Abstract: A method and system for executing a query in parallel is disclosed. A master node may receive a query from a client and develop query plans from that query. The query plans may be forwarded to worker nodes for execution, and each query plan may be accompanied by query metadata. The metadata may be stored in a catalog on the master node.

Abstract: A method and system for executing database queries in parallel using a shared metadata store. The metadata store may reside on a master node, and the master node may distribute this metadata with query plans to worker nodes. The worker nodes may request additional metadata from the master node when necessary.

Abstract: A method and system for executing a query in parallel is disclosed. A master node may receive a query from a client and develop query plans from that query. The query plans may be forwarded to worker nodes for execution, and each query plan may be accompanied by query metadata. The metadata may be stored in a catalog on the master node.

Abstract: A method and system for executing a query in parallel is disclosed. A master node may receive a query from a client and develop query plans from that query. The query plans may be forwarded to worker nodes for execution, and each query plan may be accompanied by query metadata. The metadata may be stored in a catalog on the master node.

Abstract: A method and system for executing database queries in parallel using a shared metadata store. The metadata store may reside on a master node, and the master node may distribute this metadata with query plans to worker nodes. The worker nodes may request additional metadata from the master node when necessary.

Abstract: A method and data structure is described that builds summary information using processor time that is usually proportional to the size of input data and a depth of hierarchies for a plurality of attributes. The output of computation is stored in a smaller area by eliminating redundant storage and computation. An index is generated which includes tuples or rows that include lower bound values for each of the attributes, values of children of the lower bound values based on the hierarchies of the attributes, and coverage class indicators of the coverage classes of the children.

Abstract: Database system comprising nodes configured in a tree structure is disclosed. The system includes a shared metadata store on the root node. Child nodes may request metadata from their ancestors. Parents will forward the request upward until the metadata is found or the root node is reached.

Abstract: A method, system, and process for representing a decision tree in a tabular format is discussed. The format may contain all the necessary information to traverse the nodes in parallel on a distributed system while consuming an efficient amount of resources. In some embodiments, the tree may be stored in a relational database as a table.

Abstract: A method, system, and process for generating attribute class statistics for a decision tree is discussed. In an embodiment the ACS's are calculated in parallel on a distributed architecture, and may consume a constant amount of resources.

Abstract: A method, system, and process for classifying a dataset using a decision tree is discussed. The decision tree may be stored in a novel format on a distributed architecture, and the dataset classification may be performed in parallel on the distributed architecture.

Abstract: A random number generation process generated uncorrelated random numbers from identical random number sequences on parallel processing database segments of an MPP database without communications between the segments by establishing a different starting position in the sequence on each segment using an identifier that is unique to each segment, query slice information and the number of segments. A master node dispatches a seed value to initialize the random number sequence generation on all segments, and dispatches the query slice information and information as to the number of segments during a normal query plan dispatch process.

Abstract: A method for upgrading software in a shared data store comprises a reliable, safe in-place method of transformation of metadata in which a metadata catalog is cloned, stored in the data store, and the cloned metadata is transformed in-place using the data store engine and SQL declarative language. The original and transformed cloned catalog files are swapped atomically, and the system upgrade is validated using the new version of software. If errors are encountered during the upgrade, the original catalog can be swapped back and the system returned to its original state.

Abstract: A random number generation process generated uncorrelated random numbers from identical random number sequences on parallel processing database segments of an MPP database without communications between the segments by establishing a different starting position in the sequence on each segment using an identifier that is unique to each segment, query slice information and the number of segments. A master node dispatches a seed value to initialize the random number sequence generation on all segments, and dispatches the query slice information and information as to the number of segments during a normal query plan dispatch process.

Abstract: Predicate abduction involves identifying, from a database statement, predicates on attributes of dimensional data, and applying those predicates to corresponding measure data so that only measure values are fetched and returned for dimension values that satisfy the predicate. In contrast to processing joins of relational data, the subset of the measure data is determined without needing to join a fact table with a dimension table and, therefore, without processing every fact record to determine which measure values are associated with records that satisfy the condition. Buffer reuse involves storing in buffers the “domains” of queries executing on multidimensional data. Responsive to detecting that execution of a database statement requires multiple iterations of extracting the same multidimensional data, the data is extracted during the first iteration and stored in a buffer so that, for each subsequent iteration, the buffer-stored data is accessed rather than re-extracting the data.

Abstract: In processing a query on multidimensional data in a multidimensional schema, a multidimensional database server performs various processes to limit the amount of data that is extracted from the database and presented to a relational database server in response to the request. A subset of data is identified to extract from the multidimensional data, based on the query. In various embodiments, cell-filtering, measure-filtering, and column-filtering criteria identified from the query are used to filter the subset of data, in order to return a minimized set of multidimensional data values from the subset of data. Therefore, computational resources are conserved with respect to further processing by the relational database server in response to the query.

Abstract: A method and data structure is described that builds summary information using processor time that is usually proportional to the size of input data and a depth of hierarchies for a plurality of attributes. The output of computation is stored in a smaller area by eliminating redundant storage and computation. An index is generated which includes tuples or rows that include lower bound values for each of the attributes, values of children of the lower bound values based on the hierarchies of the attributes, and coverage class indicators of the coverage classes of the children.

Abstract: Predicate abduction involves identifying, from a database statement, predicates on attributes of dimensional data, and applying those predicates to corresponding measure data so that only measure values are fetched and returned for dimension values that satisfy the predicate. In contrast to processing joins of relational data, the subset of the measure data is determined without needing to join a fact table with a dimension table and, therefore, without processing every fact record to determine which measure values are associated with records that satisfy the condition. Buffer reuse involves storing in buffers the “domains” of queries being executed on multidimensional data. In response to detecting that execution of a database statement requires multiple iterations of extracting the same multidimensional data, the data is extracted during the first iteration and stored in a buffer so that, for each subsequent iteration, the data in the buffer is accessed rather than again extracting the data.

Abstract: In processing a query on multidimensional data in a multidimensional schema, a multidimensional database server performs various processes to limit the amount of data that is extracted from the database and presented to a relational database server in response to the request. A subset of data is identified to extract from the multidimensional data, based on the query. In various embodiments, cell-filtering, measure-filtering, and column-filtering criteria identified from the query are used to filter the subset of data, in order to return a minimized set of multidimensional data values from the subset of data. Therefore, computational resources are conserved with respect to further processing by the relational database server in response to the query.