Abstract: Aspects described herein implement a usage-driven indexing scheme on a spreadsheet-based workbook. One aspect describes a system for implementing an indexing scheme in a spreadsheet workbook. The system comprises a processor configured to identify a plurality of cells stored in the plurality of tables, identify one or more table columns of the plurality of tables to index based on a usage parameter identified from a previously received query, generate an index for individual table columns of the identified table columns, receive the query including an operator and an identifier for at least one of the table columns, determine that the at least one of the table columns identified in the query has an index associated therewith, identify the index associated with the at least one of the table columns identified in the query, and identify a result to the query based on the identified index and the operator.

Abstract: Systems and methods are described to generate reverse dependency lists for cells in a workbook data store. In one aspect, a system populates, for a first cell in the data store, a reverse dependency list identifying a second cell, wherein the second cell comprises a second cell value that depends, at least in part, on a first cell value of the first cell. The system then detects a change to cell information for the first cell, and based on the detected change to the cell information: marks the second cell for recalculation based on the second cell being identified in the reverse dependency list for the first cell and the detected change to the cell information. For each marked cell, the system performs recalculation on the marked cell, performs recalculation recursively on any cells on which the marked cell depends, and stores recalculated cell values in a recalculated cache.

Abstract: Aspects are described to generate indexing on columns including links to other columns in a different table. One aspect describes a computer-implemented method comprising identifying, for a workbook comprising a plurality of cells and tables stored in a spreadsheet structure, a query comprising a formula, parsing the formula to identify, as one operand, a heterogeneous column of links in a first table of the plurality of tables to a column in a second table of the plurality of tables and at least one operator, automatically generating a virtual column for the heterogeneous column, the virtual column being inaccessible to a user, storing the generated virtual column, evaluating the function of the virtual column for each row of the first table, generating an inaccessible virtual index for the virtual column, storing the virtual index, and evaluating the formula based on the virtual index to generate a result.

Abstract: Aspects described herein implement a data-driven indexing scheme on a spreadsheet-based workbook. One aspect describes a system for implementing an indexing scheme in a spreadsheet workbook. The system comprises a data store, a user interface, and a processor configured to identify a plurality of cells stored in the plurality of tables, analyze individual cells of the plurality of cells to identify at least one formula or expression therein that involve table columns of the plurality of tables stored in the data store, identify the table columns based on parsing the identified at least one formula or expression, generate an index for individual table columns of the identified table columns, receive the query from the user interface, the query including an operator and an identifier for at least one of the table columns, identify and return a result to the query based on the identified index and the operator.

Abstract: Systems, methods, and software technology for provisioning end-points in a communication service with geographically relevant telephone numbers drawn from a data table that may be searched using partition keys derived from identifying characteristics of a location. In an implementation, the telephone numbers in the data table may be associated with partition keys related to their location. Then, when requests are received for available numbers in a particular location, the data table may be queried using the partition key associated with that location.