Subscribe to Posts by Email

Subscriber Count

    696

Disclaimer

All information is offered in good faith and in the hope that it may be of use for educational purpose and for Database community purpose, but is not guaranteed to be correct, up to date or suitable for any particular purpose. db.geeksinsight.com accepts no liability in respect of this information or its use. This site is independent of and does not represent Oracle Corporation in any way. Oracle does not officially sponsor, approve, or endorse this site or its content and if notify any such I am happy to remove. Product and company names mentioned in this website may be the trademarks of their respective owners and published here for informational purpose only. This is my personal blog. The views expressed on these pages are mine and learnt from other blogs and bloggers and to enhance and support the DBA community and this web blog does not represent the thoughts, intentions, plans or strategies of my current employer nor the Oracle and its affiliates or any other companies. And this website does not offer or take profit for providing these content and this is purely non-profit and for educational purpose only. If you see any issues with Content and copy write issues, I am happy to remove if you notify me. Contact Geek DBA Team, via geeksinsights@gmail.com

Pages

More nested loops in 11g : A change in nested loops around in 11g

Original and New Implementation for Nested Loop Joins

Oracle Database 11g Release 1 (11.1) introduces a new implementation for nested loop joins. As a result, execution plans that include nested loops might appear different than they did in previous releases of Oracle Database. Both the new implementation and the original implementation for nested loop joins are possible in Oracle Database 11g Release 1 (11.1). So, when analyzing execution plans, it is important to understand that the number of NESTED LOOPS join row sources might be different.
Original Implementation for Nested Loop Joins

Consider the following query:

SELECT e.first_name, e.last_name, e.salary, d.department_name
    FROM hr.employees e, hr.departments d
    WHERE d.department_name IN ('Marketing', 'Sales')
      AND e.department_id = d.department_id;

before Oracle Database 11g Release 1 (11.1), the execution plan for this query might appear similar to the following execution plan:

-------------------------------------------------------------------------------------------------
| Id  | Operation                   | Name              | Rows  | Bytes | Cost (%CPU)| Time     |
-------------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT            |                   |    19 |   722 |     3  (0)| 00:00:01 |
|   1 |  TABLE ACCESS BY INDEX ROWID| EMPLOYEES         |    10 |   220 |     1  (0)| 00:00:01 |
|   2 |   NESTED LOOPS              |                   |    19 |   722 |     3  (0)| 00:00:01 |
|*  3 |    TABLE ACCESS FULL        | DEPARTMENTS       |     2 |    32 |     2  (0)| 00:00:01 |
|*  4 |    INDEX RANGE SCAN         | EMP_DEPARTMENT_IX |    10 |       |     0  (0)| 00:00:01 |
-------------------------------------------------------------------------------------------------
 
Predicate Information (identified by operation id):
---------------------------------------------------
   3 - filter("D"."DEPARTMENT_NAME"='Marketing' OR "D"."DEPARTMENT_NAME"='Sales')
   4 - access("E"."DEPARTMENT_ID"="D"."DEPARTMENT_ID")

In this example,

  • the outer side of the join consists of a scan of the hr.departments table that returns the rows that match the condition department_name IN ('Marketing', 'Sales').
  • The inner loop retrieves the employees in the hr.employees table that are associated with those departments.

New Implementation for Nested Loop Joins
 
Oracle Database 11g Release 1 (11.1) introduces a new implementation for nested loop joins to reduce overall latency for physical I/O. When an index or a table block is not in the buffer cache and is needed to process the join, a physical I/O is required. In Oracle Database 11g Release 1 (11.1), Oracle Database can batch multiple physical I/O requests and process them using a vector I/O instead of processing them one at a time. As part of the new implementation for nested loop joins, two NESTED LOOPS join row sources might appear in the execution plan where only one would have appeared in prior releases. In such cases, Oracle Database allocates one NESTED LOOPS join row source to join the values from the table on the outer side of the join with the index on the inner side. A second row source is allocated to join the result of the first join, which includes the rowids stored in the index, with the table on the inner side of the join.

Consider the query in "Original Implementation for Nested Loop Joins". In Oracle Database 11g Release 1 (11.1), with the new implementation for nested loop joins, the execution plan for this query might appear similar to the following execution plan:

------------------------------------------------------------------------------------------------
| Id  | Operation                    | Name              | Rows  | Bytes | Cost(%CPU)| Time      |
-------------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT             |                   |    19 |   722 |     3   (0)| 00:00:01 |
|   1 |  NESTED LOOPS                |                   |       |       |            |          |
|   2 |   NESTED LOOPS               |                   |    19 |   722 |     3   (0)| 00:00:01 |
|*  3 |    TABLE ACCESS FULL         | DEPARTMENTS       |     2 |    32 |     2   (0)| 00:00:01 |
|*  4 |    INDEX RANGE SCAN          | EMP_DEPARTMENT_IX |    10 |       |     0   (0)| 00:00:01 |
|   5 |   TABLE ACCESS BY INDEX ROWID| EMPLOYEES         |    10 |   220 |     1   (0)| 00:00:01 |
-------------------------------------------------------------------------------------------------
 
Predicate Information (identified by operation id):
---------------------------------------------------
   3 - filter("D"."DEPARTMENT_NAME"='Marketing' OR "D"."DEPARTMENT_NAME"='Sales')
   4 - access("E"."DEPARTMENT_ID"="D"."DEPARTMENT_ID")

In this case, the rows from the hr.departments table constitute the outer side of the first join.

  • The inner side of the first join is the index emp_department_ix.
  • The results of the first join constitute the outer side of the second join, which has the hr.employees table as its inner side.

There are cases where a second join row source is not allocated, and the execution plan looks the same as it did in prior releases. The following list describes such cases:

  • All of the columns needed from the inner side of the join are present in the index, and there is no table access required. In this case, Oracle Database allocates only one join row source.
  • The order of the rows returned might be different than it was in previous releases.
  • Hence, when Oracle Database tries to preserve a specific ordering of the rows, for example to eliminate the need for an ORDER BY sort, Oracle Database might use the original implementation for nested loop joins.
  • The OPTIMIZER_FEATURES_ENABLE initialization parameter is set to a release before Oracle Database 11g Release 1 (11.1). In this case, Oracle Database uses the original implementation for nested loop joins.

-Thanks

Geek DBA

Source: Oracle 11g Documentation

Comments are closed.