Abstract: A technique includes storing, by a computer, a composite metadata object in a catalog. The composite metadata object represents information about a plurality of database objects of a table structure; the table structure includes a plurality of rows and a plurality of columns; the plurality of database objects is associated with a subset of rows of the plurality of rows; and the plurality of database objects is associated with a plurality of files corresponding to the columns. The technique includes processing, by the computer, a database transaction based on the based on composite metadata object.

Abstract: A non-transitory machine-readable storage medium stores instructions that upon execution cause a processor to, in response to a receipt of a query, communicate, via a first encrypted channel using a first key, first encrypted data between a plurality of nodes of a distributed database system, the first encrypted data comprising query metadata and a second key; communicate, via a second encrypted channel, second encrypted data between the plurality of nodes of the distributed database system, the second encrypted channel using the second key included in the first encrypted data, the second encrypted data comprising query data; and generate, by the plurality of nodes of the distributed database system, results for the query based on the query data in the second encrypted data.

Abstract: In an example, a top-k function is associated with a top-k projection for a data storage system. Input data to be loaded into the data storage system is divided and ranked according to the top-k function and stored in the top-k projection.

Abstract: A technique includes storing, by a computer, a composite metadata object in a catalog. The composite metadata object represents information about a plurality of database objects of a table structure; the table structure includes a plurality of rows and a plurality of columns; the plurality of database objects is associated with a subset of rows of the plurality of rows; and the plurality of database objects is associated with a plurality of files corresponding to the columns. The technique includes processing, by the computer, a database transaction based on the based on composite metadata object.

Abstract: According to an example, in a method for managing storage of data from an input table, a projection definition that includes an expression to be executed on data contained in a column of the input table may be accessed. The expression on the data contained in the column of the input table may be executed to obtain results data, which may be stored in an expression projection, in which the expression projection includes a column that provides physical storage for the results data. The results data may be stored in the expression projection, segmented, and encoded. In addition, the expression projection may be stored in a node.

Abstract: In an example, an aggregate function is associated with an aggregate projection for a data storage system. Input data to be loaded into the data storage system is aggregated according to the aggregate function and stored in the aggregate projection.

Abstract: A non-transitory machine-readable storage medium stores instructions that upon execution cause a processor to, in response to a receipt of a query, communicate, via a first encrypted channel using a first key, first encrypted data between a plurality of nodes of a distributed database system, the first encrypted data comprising query metadata and a second key; communicate, via a second encrypted channel, second encrypted data between the plurality of nodes of the distributed database system, the second encrypted channel using the second key included in the first encrypted data, the second encrypted data comprising query data and generate, by the plurality of nodes of the distributed database system, results for the query based on the query data in the second encrypted data.

Abstract: In an example, an aggregate function is associated with an aggregate projection for a data storage system. Input data to be loaded into the data storage system is aggregated according to the aggregate function and stored in the aggregate projection.

Abstract: According to an example, in a method for managing storage of data from an input table, a projection definition that includes an expression to be executed on data contained in a column of the input table may be accessed. The expression on the data contained in the column of the input table may be executed to obtain results data, which may be stored in an expression projection, in which the expression projection includes a column that provides physical storage for the results data. The results data may be stored in the expression projection, segmented, and encoded. In addition, the expression projection may be stored in a node.

Abstract: In an example, a top-k function is associated with a top-k projection for a data storage system. Input data to be loaded into the data storage system is divided and ranked according to the top-k function and stored in the top-k projection.

Abstract: In an example, data from a source location is merged into a target location, in which both the source location and the target location contain tuples of data. A matched tuple in the target table that matches a The deleting of the tuple in the target table includes executing a semi inner join operation between the source table and the target table to identify the matched tuple. Each tuple contained in the source table may be inserted into the target table as a new tuple.

Abstract: According to an example, a method for executing a query having multiple set operators includes adding an additional column that associates an identifier of the result table with data contained in the input result table in each input result table of an input query. The method also includes performing a Union All operation on the data contained in the input result tables to generate an intermediate result set, performing a Group By operation on the data contained in the intermediate result set to generate a grouped result set, wherein the Group By operation operates on various groups of rows and returns one row per each of the various groups, adding aggregation columns in the grouped result set containing counts of a number of times that each tuple exists in each of the input result tables, and applying a predicate on the grouped result set to execute the query.