Pruning disk blocks in a relational database management system

• US 10,223,422 B2
• Filed: 03/27/2015
• Issued: 03/05/2019
• Est. Priority Date: 09/28/2012
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Ziauddin, Mohamed, Witkowski, Andrew
• Primary Examiner(s): Pham, Khanh B
• Application Number:
  US14/671,510
• Publication Number:
  US 20150199407A1
• Time in Patent Office: 1,439 Days
• CPC Class Codes: G06F 16/24557 Efficient disk access during query execution
• CPC Class Codes: G06F 16/278 Data partitioning, e.g. horizontal or vertical partitioning
• CPC Class Codes: G06F 16/283 Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP
• CPC Class Codes: G06F 16/285 Clustering or classification

Abstract

Techniques are provided for generating a “dimensional zonemap” that allows a database server to avoid scanning disk blocks of a fact table based on filter predicates in a query that qualify one or more dimension tables. The zonemap divides the fact table into sets of contiguous disk blocks referred to as “zones”. For each zone, a minimum value and a maximum value for each of one or more “zoned” columns of the dimension tables is determined and maintained in the zonemap. For a query that contains a filter predicate on a zoned column, the predicate value can be compared to the minimum value and maximum value maintained for a zone for that zoned column to determine whether a scan of the disk blocks of the zone can be skipped.

Claims

1. A computer-implemented method comprising:

dividing disk blocks of a target table into a plurality of zones, each zone of the plurality of zones being defined by a respective set of contiguous disk blocks on disk, the respective set of contiguous disk blocks on disk containing data of one or more rows of the target table;

determining, for each zone of the plurality of zones, a minimum column value and a maximum column value for each of one or more zoned columns of one or more tables;

wherein each zone of the plurality of zones corresponds to a respective set of data in the one or more zoned columns;

wherein, for each zone of the plurality of zones, the minimum column value and maximum column value determined for each zoned column of the one or more zoned columns is determined based on the respective set of data, in the one or more zoned columns, corresponding to the zone;

wherein at least one of the one or more zoned columns is not a column of target table;

receiving a query on the target table;

wherein the query contains a filter predicate on a particular column of the one or more zoned columns;

wherein the filter predicate has a predicate value;

determining, based on the predicate value and the minimum column values and the maximum column values maintained for the particular column, one or more zones, of the plurality of zones, that cannot possibly contain rows of the target table satisfying the query;

wherein the method is performed by one or more computing devices.

2. The method of claim 1, further comprising:

returning an answer to the query without performing a disk scan of the one or more zones, of the plurality of zones, that cannot possibly contain rows of the target table satisfying the query.

3. The method of claim 1, wherein the particular column is not a column of the target table.

4. The method of claim 1, wherein none of the one or more zoned columns is a column of the target table;

and wherein the one or more zoned columns is a plurality of zoned columns.

5. The method of claim 1, wherein the target table is a fact table of a star schema and the particular column is a column of a dimension table of the star schema.

6. The method of claim 1, wherein the particular column is not a column of the target table, and wherein dividing disk blocks of the target table into a plurality of zones comprises:

performing a left outer join between the target table and the table having the particular column to produce a set of joined rows, each joined row of the set of joined rows corresponding to one row of the target table; and

for each joined row in the set of joined rows, assigning the corresponding row of the target table to one zone of the plurality of zones.

7. The method of claim 1, wherein each zone of the plurality of zones has at most a predetermined number of contiguous disk blocks.

8. The method of claim 1, wherein maintaining, for each zone of the plurality of zones, a minimum column value and a maximum column value for each of the one or more zoned columns comprises maintaining a materialized view in a database, the materialized view including one row for each zone of the plurality of zones, one column for each of the one or more zoned columns storing the minimum column values for the zoned column, and one column for each of the one or more columns storing the maximum column values for the zoned column.

9. The method of claim 1, wherein the filter predicate contains a LIKE operator and the predicate value has a constant character string value or a constant character string value prefixing a wildcard operator.

10. The method of claim 1, wherein the particular column is not a column of the target table, and wherein the query contains join criteria, the join criteria specifying a join between the target table and the table having the particular column, the join criteria further specifying that the join is to be based on a foreign key column of the target table and a unique key column of the table having the particular column.

11. The method of claim 1, wherein the particular column has a database data type;

and wherein the database data type is selected from the group consisting of numeric, character, date, timestamp, row identifier, float, and double.

12. One or more non-transitory computer-readable media storing at least one program configured for execution by at least one processor, the at least one program comprising instructions to:

dividing disk blocks of a target table into a plurality of zones, each zone of the plurality of zones being defined by a respective set of contiguous disk blocks on disk, the respective set of contiguous disk blocks on disk containing data of one or more rows of the target table;

determining, for each zone of the plurality of zones, a minimum column value and a maximum column value for each of one or more zoned columns of one or more tables;

wherein each zone of the plurality of zones corresponds to a respective set of data in the one or more zoned columns;

wherein, for each zone of the plurality of zones, the minimum column value and maximum column value determined for each zoned column of the one or more zoned columns is determined based on the respective set of data, in the one or more zoned columns, corresponding to the zone;

wherein at least one of the one or more zoned columns is not a column of target table;

receive a query on the target table;

wherein the query contains a filter predicate on a particular column of the one or more zoned columns;

wherein the filter predicate has a predicate value;

determine, based on the predicate value and the minimum column values and the maximum column values maintained for the particular column, one or more zones, of the plurality of zones, that cannot possibly contain rows of the target table satisfying the query.

13. The one or more non-transitory computer-readable media of claim 12, the at least one program further comprising instructions to:

return an answer to the query without performing a disk scan of the one or more zones, of the plurality of zones, that cannot possibly contain rows of the target table satisfying the query.

14. The one or more non-transitory computer-readable media of claim 12, wherein the particular column is not a column of the target table.

15. The one or more non-transitory computer-readable media of claim 12, wherein none of the one or more zoned columns is a column of the target table;

and wherein the one or more zoned columns is a plurality of zoned columns.

16. The one or more non-transitory computer-readable media of claim 12, wherein the target table is a fact table of a star schema and the particular column is a column of a dimension table of the star schema.

17. The one or more non-transitory computer-readable media of claim 12, wherein the particular column is not a column of the target table, and wherein the instructions to divide contiguous disk blocks of the target table into a plurality of zones comprise instructions to:

perform a left outer join between the target table and the table having the particular column to produce a set of joined rows, each joined row of the set of joined rows corresponding to one row of the target table; and

for each joined row in the set of joined rows, assign the corresponding row of the target table to one zone of the plurality of zones.

18. The one or more non-transitory computer-readable media of claim 12, wherein each zone of the plurality of zones has at most a predetermined number of contiguous disk blocks.

19. The one or more non-transitory computer-readable media of claim 12, wherein the instructions to maintain, for each zone of the plurality of zones, a minimum column value and a maximum column value for each of the one or more zoned columns comprise instructions to maintain a materialized view in a database, the materialized view including one row for each zone of the plurality of zones, one column for each of the one or more zoned columns storing the minimum column values for the zoned column, and one column for each of the one or more columns storing the maximum column values for the zoned column.

20. The one or more non-transitory computer-readable media of claim 12, wherein the filter predicate contains a LIKE operator and the predicate value has a constant character string value or a constant character string value prefixing a wildcard operator.

21. The one or more non-transitory computer-readable media of claim 12, wherein the particular column is not a column of the target table, and wherein the query contains join criteria, the join criteria specifying a join between the target table and the table having the particular column, the join criteria further specifying that the join is to be based on a foreign key column of the target table and a unique key column of the table having the particular column.

22. The one or more non-transitory computer-readable media of claim 12, wherein the particular column has a database data type;

and wherein the database data type is selected from the group consisting of numeric, character, date, timestamp, row identifier, float, and double.