Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
2 views

SQL-30-Performance-Tips-_-Cheat-Sheet

This document provides 30 performance tips for SQL, categorized into sections such as fetching, filtering, aggregating, joining data, and indexing. Key recommendations include selecting only necessary columns, using appropriate join types, and maintaining indexes effectively. The tips aim to enhance query performance and efficiency in database management.

Uploaded by

postbox181
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
2 views

SQL-30-Performance-Tips-_-Cheat-Sheet

This document provides 30 performance tips for SQL, categorized into sections such as fetching, filtering, aggregating, joining data, and indexing. Key recommendations include selecting only necessary columns, using appropriate join types, and maintaining indexes effectively. The tips aim to enhance query performance and efficiency in database management.

Uploaded by

postbox181
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

SQL 30 Performance Tips

Always Test using Execution Plan


CHEAT SHEET

FETCHING DATA AGGREGATING DATA


#1 Only select the columns you need. Avoid using SELECT * #17 Use columnstore indexes for queries involving heavy aggregations on large tables

#2 Avoid using DISTINCT or ORDER BY unless absolutely necessary, as they can slow down queries #18 Pre-aggregate data and store the results in a separate table for faster reporting

#3 For exploration, limit the number of rows using TOP to avoid fetching unnecessary data

SUBQUERIES
FILTERING DATA #19 Understand when to use JOIN, EXISTS, or IN. Avoid IN with large lists as it can be inefficient

#4 Create non-clustered indexes on columns frequently used in the WHERE to speed up queries #20 Simplify your queries by eliminating redundant logic and conditions by using CTE

#5 Avoid functions e.g., UPPER(), YEAR() to columns in the WHERE , as this prevents indexes from being used

#6 Avoid starting string searches with a wildcard (%example), as this disables index usage
DDL
#7 Use IN instead of multiple OR conditions for better readability and performance
#21 Avoid VARCHAR or TEXT types unnecessarily; choose precise data types to save storage and improve performance

#22 Avoid defining excessive lengths in your data types (e.g., VARCHAR(MAX)) unless truly needed

JOINING DATA #23 Use NOT NULL constraints wherever possible to enforce data integrity

#8 Understand the performance implications of different join types. Use INNER JOIN when possible for efficiency #24 Ensure all tables have a clustered primary key to provide structure and improve query performance

#9 Always use explicit (ANSI-style) joins (INNER JOIN, LEFT JOIN, etc.) instead of older implicit join syntax #25 Add non-clustered indexes to foreign keys that are frequently queried to speed up lookups

#10 Ensure that the columns in the ON of your joins are indexed for optimal performance

#11 Filter before joining large tables to reduce the size of the dataset being joined
INDEXING
#12 Aggregate before joining large tables to reduce the size of the dataset being joined
#26 Avoid Over Indexing, as it can slow down insert, update, and delete operations

#13 Replace OR conditions in join logic with UNION where possible to improve query performance
#27 Regularly review and drop unused indexes to save space and improve write performance

#14 Be aware of nested loops in your query execution plan. Use SQL Hints if needed to optimize performance #28 Update table statistics weekly to ensure the query optimizer has the most up-to-date information

#15 Use UNION ALL instead of UNION if duplicates are acceptable, as it is faster
#29 Reorganize and rebuild fragmented indexes weekly to maintain query performance.

#16 When duplicates are not acceptable, use UNION ALL + DISTINCT instead of UNION for better performance
#30 For large tables (e.g., fact tables), partition the data and then apply a columnstore index for best performance results

You might also like