Scribd
Upload
20 views25 pages

1 Query Tuning Introduction m01 Basics Execution Plan Slides

The document discusses execution plan fundamentals, including query optimization, operator types like scans and seeks, and join methods. It explains that SQL Server's query optimizer aims to select the most efficient execution plan, with specialized plans sometimes needed. Operators are classified as generally good or bad based on an 80-20 rule with exceptions.

Uploaded by

Orcun Iyigun
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
20 views25 pages

1 Query Tuning Introduction m01 Basics Execution Plan Slides

The document discusses execution plan fundamentals, including query optimization, operator types like scans and seeks, and join methods. It explains that SQL Server's query optimizer aims to select the most efficient execution plan, with specialized plans sometimes needed. Operators are classified as generally good or bad based on an 80-20 rule with exceptions.

Uploaded by

Orcun Iyigun
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 25

Basics to Execution Plan

Pinal Dave
http://blog.sqlauthority.com

Vinod Kumar M
http://blogs.extremeexperts.com
Execution Plans are
like Weather Forecasts
Getting Started

 Query Optimization Fundamentals


 Execution Plan Fundamentals
 Execution Plan Operators
 Scan
 Seek

 Joins
Query Optimization

 Cost Based Rules


 Generate Multiple Execution Plans
 Selects Lowest Cost Execution Plans

 Maps Logical Query Operations to


Physical Operations
 Logical (Inner) Join -> Nested Loop Join
 Sort -> Sort
Finding Optimal Plan

Optimal
Plan Quality

Optimal
Plan
Selection
Time
Plan Cost

 Sum of all physical operator in plan


 CPU
 IO

 Memory
Display Options of Execution Plans

 SET STATISTICS TIME ON


 SET STATISTICS IO ON
Types of Execution Plans

 Estimated Execution Plan


 Actual Execution Plan
Display Types of Execution Plans

 Graphic
 Text
 XML
Right Click Options on Operators

 Save Execution Plan As..


 Show Execution Plan XML…
 Zoom In
 Zoom Out
 Zoom Fit
 Properties
Demo

 Query Cost
 Types of Execution Plans
 Execution Plan Properties
 Basic Building Blocks
Execution Plan Operators

 When
 Good
 Bad
 Action Item
80% - 20% Rule
Table Scan

 When
 Table without clustered index is accessed
 Good or Bad*
 Can’t decide
 Action Item
 Create Clustered Index

* 80% - 20% rule


Clustered Index Scan

 When
 Table with clustered index is accessed
 Query does not use non clustered index
 Table does not have non clustered index

 Good or Bad*
 Bad unless large data with most columns
and rows retrieved
 Action Item
 Evaluate Clustered Index Keys

* 80% - 20% rule


Clustered Index Seek

 When
 Table with clustered index is accessed and
query locates specific rows in B+ tree
 Good or Bad*
 Good
 Action Item
 Evaluate possibility of non-clustered index

* 80% - 20% rule


Non-Clustered Index Scan

 When
 Columns part of non-clustered index
accessed in query
 Good or Bad
 Bad unless large data with most columns
and rows retrieved
 Action Item
 Create more refined non-clustered index

* 80% - 20% rule


Non-Clustered Index Seek

 When
 Columns part of non-clustered index
accessed in query and rows located in B+
tree
 Good or Bad*
 Good
 Action Item
 Further evaluate other operators

* 80% - 20% rule


Lookups

 When
 Query Optimizer uses non-clustered index
to search few column data and base table
for other columns data
 Good or Bad*
 Bad
 Action Item
 Included Index or Covered Index

* 80% - 20% rule


Seek or Scan
Nested Loops Join

 When
 A nested loops join is particularly effective
if the outer input is small and the inner
input is pre-indexed and large
 Good or Bad*
 Good in small transactions
 Action Item
 Optimizer knows best

* 80% - 20% rule


Merge Join

 When
 A merge loops join is particularly effective
when both the inputs are sorted on the
merge column
 Good or Bad*
 Good when query has large result set
 Bad when inputs are not sorted

 Action Item
 Optimizer knows best
* 80% - 20% rule
Hash Join

 When
 A hash join is particularly effective with
large data requires many types of set-
matching operations
 Good or Bad*
 Good when query requires set-matching
operations
 Action Item
 Optimizer knows best
* 80% - 20% rule
Optimal Join
Optimizer’s Choice
Summary

 Query Optimization Fundamentals


 Execution Plan Fundamentals
 Execution Plan Operators
 Scan
 Seek

 Joins

Remember: SQL Server Optimizer usually opts for most efficient execution plan.

Remember: 80%-20% Rule. There are always special cases.

You might also like