Abstract: Techniques are provided for autonomous caching of hierarchical data. In one technique, query log data is stored that comprises multiple entries, each entry (a) corresponding to a different instance of a query that was executed against a database and (b) indicating a tuple (or level grouping) comprising a set of hierarchy levels corresponding to a set of dimensions. Multiple tuples indicated in the query log data are identified. For each tuple: (1) a set of entries that indicate the tuple is identified and the set of entries is associated with the tuple; (2) aggregated performance data is generated for the tuple based on performance data associated with each entry in the set of entries that is associated with the tuple; and (3) based on the aggregated performance data, it is determined whether to create a new auto-cache table or to delete an existing auto-cache table associated with the tuple.

Abstract: Herein are techniques for automatically leveraging metadata of an analytic view to accelerate a relational query. In an embodiment, a computer stores model metadata that defines an analytic view that contains a join operation that is based on a dimension column of a dimension table and a join column of a fact table. The analytic view also contains a measure that is based on an aggregation operation and a measure column of the fact table. Also stored is denormalization metadata that defines a transparency view that is based on the analytic view. In operation, a query that references the transparency view is received. The query does not reference the analytic view. The query that references the transparency view is executed based on: a) the denormalization metadata that defines the transparency view, b) the model metadata that defines the analytic view, and c) the measure that is based on the aggregation operation and the measure column of the fact table.

Abstract: Herein are techniques for automatically leveraging metadata of an analytic view to accelerate a relational query. In an embodiment, a computer stores model metadata that defines an analytic view that contains a join operation that is based on a dimension column of a dimension table and a join column of a fact table. The analytic view also contains a measure that is based on an aggregation operation and a measure column of the fact table. Also stored is denormalization metadata that defines a transparency view that is based on the analytic view. In operation, a query that references the transparency view is received. The query does not reference the analytic view. The query that references the transparency view is executed based on: a) the denormalization metadata that defines the transparency view, b) the model metadata that defines the analytic view, and c) the measure that is based on the aggregation operation and the measure column of the fact table.

Abstract: Techniques are provided for autonomous caching of hierarchical data. In one technique, query log data is stored that comprises multiple entries, each entry (a) corresponding to a different instance of a query that was executed against a database and (b) indicating a tuple (or level grouping) comprising a set of hierarchy levels corresponding to a set of dimensions. Multiple tuples indicated in the query log data are identified. For each tuple: (1) a set of entries that indicate the tuple is identified and the set of entries is associated with the tuple; (2) aggregated performance data is generated for the tuple based on performance data associated with each entry in the set of entries that is associated with the tuple; and (3) based on the aggregated performance data, it is determined whether to create a new auto-cache table or to delete an existing auto-cache table associated with the tuple.

Abstract: Techniques related to group determination based on multi-table dictionary codes are disclosed. In some embodiments, one or more non-transitory storage media store a sequence of instructions which, when executed by one or more computing devices, cause performance of a method. The method comprises storing a fact table and a dimension table that share a domain dictionary. The fact table and the dimension table each have a column of encoded join keys that is decodable using the shared domain dictionary. A query may specify one or more row groups for the dimension table. To efficiently process the query, one or more group identifiers are assigned to the one or more row groups. Each row group corresponds to a different group identifier. This enables a code-to-group-identifier mapping to be generated. The code-to-group-identifier mapping correlates the encoded join keys to the one or more group identifiers.

Abstract: Techniques related to cache storage formats are disclosed. In some embodiments, a set of values is stored in a cache as a set of first representations and a set of second representations. For example, the set of first representations may be a set of hardware-level representations, and the set of second representations may be a set of non-hardware-level representations. Responsive to receiving a query to be executed over the set of values, a determination is made as to whether or not it would be more efficient to execute the query over the set of first representations than to execute the query over the set of second representations. If the determination indicates that it would be more efficient to execute the query over the set of first representations than to execute the query over the set of second representations, the query is executed over the set of first representations.

Abstract: Techniques related to group determination based on multi-table dictionary codes are disclosed. In some embodiments, one or more non-transitory storage media store a sequence of instructions which, when executed by one or more computing devices, cause performance of a method. The method comprises storing a fact table and a dimension table that share a domain dictionary. The fact table and the dimension table each have a column of encoded join keys that is decodable using the shared domain dictionary. A query may specify one or more row groups for the dimension table. To efficiently process the query, one or more group identifiers are assigned to the one or more row groups. Each row group corresponds to a different group identifier. This enables a code-to-group-identifier mapping to be generated. The code-to-group-identifier mapping correlates the encoded join keys to the one or more group identifiers.

Abstract: Methods, computer systems, and stored instructions are described herein for densely grouping dimensional data and/or aggregating data using a data structure, such as one that is constructed based on dimensional data. When smaller tables are joined with a larger table, a server may analyze the smaller tables first to determine actual value combinations that occur in the smaller tables, and these actual value combinations are used to more efficiently process the larger table. A dense data structure may be generated by processing dimensional data before processing data from fact table. The dense data structure may be generated by compressing ranges of values that are possible in dimensions into a range of values that actually occurs in the dimensions. The compressed range of values may be represented by dense set identifiers rather than the actual compressed range of values.

Abstract: Methods, computer systems, and stored instructions are described herein for densely grouping dimensional data and/or aggregating data using a data structure, such as one that is constructed based on dimensional data. When smaller tables are joined with a larger table, a server may analyze the smaller tables first to determine actual value combinations that occur in the smaller tables, and these actual value combinations are used to more efficiently process the larger table. A dense data structure may be generated by processing dimensional data before processing data from fact table. The dense data structure may be generated by compressing ranges of values that are possible in dimensions into a range of values that actually occurs in the dimensions. The compressed range of values may be represented by dense set identifiers rather than the actual compressed range of values.

Abstract: Methods, computer systems, and stored instructions are described herein for transforming a query or execution plan to re-use stored data. A query processor stores a query or a representation thereof. Before executing the query, the query processor analyzes the query or representation to determine whether the query or representation could cause at least part of a data structure to be generated at least twice. Based at least in part on determining that the at least part of the data structure could be generated at least twice by the query or representation, the query processor transforms the query or representation. The transformed query or representation includes a first transformed sub-query or sub-operation that generates and stores the at least part of the data structure, and a second transformed sub-query or sub-operation uses the at least part of the data structure that was stored by the first transformed sub-query or sub-operation.

Abstract: Methods, computer systems, and stored instructions are described herein for densely grouping dimensional data and/or aggregating data using a data structure, such as one that is constructed based on dimensional data. When smaller tables are joined with a larger table, a server may analyze the smaller tables first to determine actual value combinations that occur in the smaller tables, and these actual value combinations are used to more efficiently process the larger table. A dense data structure may be generated by processing dimensional data before processing data from fact table. The dense data structure may be generated by compressing ranges of values that are possible in dimensions into a range of values that actually occurs in the dimensions. The compressed range of values may be represented by dense set identifiers rather than the actual compressed range of values.

Abstract: Methods, computer systems, and stored instructions are described herein for densely grouping dimensional data and/or aggregating data using a data structure, such as one that is constructed based on dimensional data. When smaller tables are joined with a larger table, a server may analyze the smaller tables first to determine actual value combinations that occur in the smaller tables, and these actual value combinations are used to more efficiently process the larger table. A dense data structure may be generated by processing dimensional data before processing data from fact table. The dense data structure may be generated by compressing ranges of values that are possible in dimensions into a range of values that actually occurs in the dimensions. The compressed range of values may be represented by dense set identifiers rather than the actual compressed range of values.