# US 2013/0159245 A1 · Data extraction and sparseness discovery from multidimensional databases

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2013/0159245 A1
- Document ID: US20130159245A1
- Status: Published application
- Kind: A1
- Application number: 13/711,344
- Filed: 2012-12-11
- Published: 2013-06-20
- Earliest claimed: 2011-12-16
- Assignees: Star Analytics, Inc.
- Inventors: Steffen Zacho Frederiksen
- Web page: https://patentgrep.com/patent/US20130159245A1

## CPC

- G06F 16/283: {Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP}
- G06F 16/2264: {Multidimensional index structures}
- G06F 16/254: {Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Techniques for a method for efficiently extracting data from multidimensional databases are disclosed herein. When extracting data from a subset of a multidimensional database, the method first extracts data from high-level cells within the subset. If there are any empty high-level cells, the method removes lower-level cells that roll-up into the empty high-level cell, without extracting data from these lower-level cells. The method improves the data extraction efficiency by reducing the overhead of extracting data from empty cells of multidimensional databases.

## Claims

### Claim 1

A method comprising:
receiving a request to extract data from a subset of a multidimensional database that stores data at a plurality of cells;
determining, from the subset, a reduced subset by excluding a plurality of empty cells from the subset; and
satisfying the request by extracting data values from only the reduced subset.

### Claim 2 (depends on 1)

The method of claim 1, wherein the subset is as having a plurality of dimension members, and each dimension member of the plurality of dimension members belongs to a dimension of a plurality of dimensions of the multidimensional database.

### Claim 3 (depends on 1)

The method of claim 1, wherein the plurality of empty cells includes dependent cells of high-level cells, each of the high-level cells has at least N−1 top-level members, and N is a total number of dimensions of the multidimensional database.

### Claim 4 (depends on 3)

The method of claim 3, wherein each of the top-level members' values does not roll up into the value of any ancestor member from the subset.

### Claim 5 (depends on 3)

The method of claim 3, wherein each of data values from the dependent cells of the high-level cells contributes to at least a portion of a data value from the high-level cells.

### Claim 6 (depends on 1)

The method of claim 1, wherein each empty cell of the plurality of empty cells has no data value stored in that empty cell.

### Claim 7 (depends on 1)

The method of claim 1, wherein the subset only includes leaf level cells, and the leaf level cells are located by leaf level dimension members.

### Claim 8 (depends on 7)

The method of claim 7, further comprising:
adding new cells into the subset, wherein each of the new cells is located by a dimension member that is not a leaf level dimension member, and each of the leaf level cells is a dependent cell of one of the new cells, wherein a data value from each of the dependent cells contributes to a data value from one of the high-level cells.

### Claim 9 (depends on 7)

The method of claim 7, wherein the leaf level cells include at least one empty cell, and the empty cell has no data value stored in the empty cell.

### Claim 10 (depends on 2)

The method of claim 2, wherein the plurality of empty cells includes empty high-level cells, and wherein high-level cells are cells having at least one top-level member, and wherein a top-level member is a dimension member, of the plurality of dimension members, whose value does not roll-up into the value of any ancestor members from the plurality of dimension members.

### Claim 11 (depends on 10)

The method of claim 10, wherein the plurality of empty cells further includes roll-up cells of the empty top-level cells, and wherein the roll-up cells of the empty top-level cells are cells whose dimension members are either descendent members of or the same dimension members as the dimension members of the empty top-level cells.

### Claim 12 (depends on 1)

The method of claim 1, wherein the step of determining comprises:
identifying all top-level members from a plurality of dimension members that define the subset, wherein the top-level members' values do not roll up into a value of any ancestor member from the plurality of dimension members;
extracting a data value from a high-level cell located by the top-level members and a group of dimension members from the plurality of dimension members; and
if the data value is an empty data value, determining, from the subset, a reduced subset by excluding roll-up cells of the high-level cell from the reduced subset, wherein each of the roll-up cells are located by a descendent member of the top-level member and the group of dimension members.

### Claim 13 (depends on 12)

The method of claim 12, wherein one of the roll-up cells is located by a leaf level dimension member.

### Claim 14 (depends on 13)

The method of claim 13, further comprising:
determining middle-level cells outside the subset, wherein each of the leaf-level cells is a roll-up cell of the middle-level cells, and a data value from the roll-up cell contributes to a data value of the middle-level cells;
for each individual middle-level cell of the middle-level cells,
extracting a data value from that individual middle-level cell,
if that individual middle-level cell has no data value, determining, from the subset, a reduced subset by excluding roll-up cells of that individual middle-level cell from the reduced subset, wherein a data value from each of the roll-up cells contributes to a data value from that individual middle-level cell.

### Claim 15

A method comprising:
receiving a request to extract data from a subset of a multidimensional database that stores data at a plurality of cells, wherein the subset is defined as having a plurality of dimension members;
identifying a top-level member from the plurality of dimension members, wherein the top-level member has no ancestor member from the subset;
extracting a data value from a cell located by a dimension member and a group of top-level members;
if the data value is an empty data value, determining, from a subset, a reduced subset by excluding dependent cells of the cell from the reduced subset, wherein the dependent cells have the dimension member and descendent members of the group of top-level members; and
satisfying the request by extracting data values from only the reduced subset.

### Claim 16 (depends on 15)

The method of claim 15, wherein the top-level member is a root dimension member.

### Claim 17 (depends on 16)

The method of claim 16, wherein each of data values from the dependent cells of the cell contributes to at least a portion of the data value from the cell.

### Claim 18 (depends on 15)

The method of claim 15, wherein the subset is defined by only a plurality of leaf level dimension members;
and the method further comprises:
adding new cells from the multidimensional database into the subset, wherein each of the new cells is located by an ancestor member of a leaf level dimension member of the plurality of leaf level dimension members.

### Claim 19 (depends on 18)

The method of claim 18, wherein the subset includes a plurality of leaf level cells that are located by the plurality of leaf level dimension members, and each of the leaf level cells is a dependent cell of one of the new cells, wherein a data value from each of the dependent cells contributes to a data value from one of the new cells.

### Claim 20

A method comprising:
receiving a request to extract data from a subset of a multidimensional database that stores data at a plurality of cells, wherein the subset is defined by a plurality of dimension members, each dimension member of the plurality of dimension members belongs to a dimension of a plurality of dimensions of the multidimensional database;
identifying top-level members from the plurality of dimension members, wherein the top-level members have no ancestor member from subset;
selecting a dimension from the plurality of dimensions, a dimension member of the selected dimension, and a cell of the subset that has the selected dimension member and top-level members of dimensions other than the selected dimension;
extracting a data value from the selected cell;
if the selected cell has no data value, determining, from the subset, a reduced subset by excluding dependent cells of the selected cell from the reduced subset, wherein each of the dependent cells has the selected dimension member and descendent members of the top-level members of dimensions other than the selected dimension; and
satisfying the request by extracting data values from the reduced subset without extracting data values from the excluded dependent cells.

### Claim 21 (depends on 20)

The method of claim 20, further comprises:
repeating the steps of extracting and determining for all cells of the subset that have the selected dimension member and top-level members of dimensions other than the selected dimension.

### Claim 22 (depends on 20)

The method of claim 20, further comprises:
repeating the steps of extracting and determining for all dimension members of the selected dimension and all cells of the subset that have the dimension members of the selected dimension and top-level members of dimensions other than the selected dimension.

### Claim 23 (depends on 20)

The method of claim 20, further comprises:
repeating the steps of extracting and determining for all cells of the subset that have at least N−1 top-level members, wherein N is a total number of dimensions of the multidimensional database.

### Claim 24

A database management system comprising:
a processor;
a data storage device to store a multidimensional database; and
a memory storing instructions which, when executed by the processor, cause the storage system to perform a process including:
receiving a request to extract data from a subset of the multidimensional database that stores data at a plurality of cells;
determining, from the subset, a reduced subset by excluding a plurality of empty cells from the reduced subset; and
satisfying the request by extracting data values from only the reduced subset.

### Claim 25 (depends on 24)

The database management system of claim 24, wherein the step of determining comprises:
Identifying a top-level member from a plurality of dimension members that define the subset, wherein the top-level member has no ancestor member from the subset;
extracting a data value from a cell from the subset located by a dimension member and a group of top-level members; and
if the data value is an empty data value, determining, from the subset, a reduced subset by excluding dependent cells of the cell from the reduced subset, wherein the dependent cells have the dimension member and descendent members of the group of top-level members.
