This document summarizes a lunch and learn session on SQL tips and tricks for developers. It covers topics like using indexes to improve performance, avoiding SQL injection, using transactions properly, null value handling, the N+1 query problem, database statistics and tools like Spring Data repositories, IntelliJ SQL support and database versioning. It also discusses NoSQL vs SQL databases and embedded database options for testing.

1. Lunch + Learn
SQL Tips + Tricks for Developers
Victor Szoltysek
July / 2016

2. Tables + Indexes

3. Common Mistakes

4. Not closing resources
• always using finally block / try-resources
finally{
sql1.close();
sql2.close();
}
@Cleanup
Resource someClosableResource;

5. Missing Indexes
• Not important for small tables
• Huge potential performance improvements — x10000
• Watch out for incorrect index usage
• MSSQL Server will recommend indexes
• ORMs don’t create indexes!!!!
SELECT *
FROM someTable
WHERE someColumn
LIKE ‘%stringInMiddle%’

7. SQL Injection

8. SQL Injection - Mitigation
• Never EVER use concatenated SQL strings in your
prod applications
• Use Parametrized Statements

9. Not using Transactions when
Needed
• Transaction :
• Logical Atomic Unit of Work
• All or nothing (even if database gets interrupted)
• Solution for A/B
• Spring @Transactional Annotation

10. Null misunderstandings!
• NULL == NULL <— not true
• NULL != NULL <— also not true
• NULL != “something” <— ALSO not true
• Empty Strings become Null in Oracle!!

11. N+1 AntiPattern
• Minimize SQL Traffic!! , Let SQL perform joins,
Network is slow
• Use SQL Count instead of getting all rows and
counting at backend
-- Get all Mobile Vendors
SELECT * FROM MobileVendor;
-- For each MobileVendor, get PhoneModel details
SELECT * FROM PhoneModel WHERE MobileVendor.vendorId=?

12. Stale Statistics
• Database keeps track of statistics about table data
to help to optimize how it executes queries
• Significant changes to data, can make these
statistics temporarily invalid, and can lead to poor
execution
• Only important with large DB changes, i.e. full
refresh

13. Tools

14. Spring Data - Auto
Generated Repositories

15. Intellij Ultimate SQL Support

16. VirtualBox + Windows + SQL Server
Express + Mangement Studio

17. Database Versioning
DBDeploy

18. Different SQL Databases
SQL Server Oracle Postgres MySQL Firebird H2
Single Cross
Platform
Deployable
No No No No No Yes
Embedded
Support
No No No No Yes Yes
InMemory
Support
No No No Yes No Yes
Free No No Yes Yes Yes Yes
Cross Platform No Yes Yes Yes Yes Yes

19. InMemory / Embedded /
Java Databases
String connectionString =
"jdbc:h2:mem:inMemoryDb;DB_CLOSE_DELAY=-1"
Awesome for Testing!!!

20. Most commonly used SQL
database?

21. Explain Plans

22. Groovy SQL Support

24. ORMs
• Great for Greenfield / CRUD apps
• Easy to go from Object Model to SQL Model
• Can get difficult to create Object Model from
existing SQL Model
• “ORMS are the Vietnam of Computer Science”

25. NoSQL vs SQL vs NewSQL
• NoSQL did not replace SQL, but is an alternative
• Relational / Structured Data is still create for SQL
DB’s
• NewSQL!! Modern SQL alternatives

26. Log4J JDBC Appender

27. SQL Metadata
SELECT
t.NAME AS TableName,
p.rows AS RowCounts,
(SUM(a.total_pages) * 8 / 1064) AS TotalSpaceMB
FROM
sys.tables t
INNER JOIN
sys.schemas s ON s.schema_id = t.schema_id
INNER JOIN
sys.indexes i ON t.OBJECT_ID = i.object_id
INNER JOIN
sys.partitions p ON i.object_id = p.OBJECT_ID AND i.index_id = p.index_id
INNER JOIN
sys.allocation_units a ON p.partition_id = a.container_id
WHERE
t.NAME NOT LIKE 'dt%' -- filter out system tables for diagramming
AND t.is_ms_shipped = 0
AND i.OBJECT_ID > 255
GROUP BY
t.Name, s.Name, p.Rows
ORDER BY
p.Rows desc

28. Table Analysis
Table Row Count Size MB
agreement_customer 1039293 103
pricing_results 378520 189
sales_history 378520 40
price_rule 256814 26
base_value 252687 19
customer 32754 3
product 19525 1
product_price 19525 5
agreement_product 15105 1
price_rule_flag 7656 0
price_rule_header 2552 0
min_max_rule 292 0
DATABASECHANGEL
OG
61 0
sub_price_rule 29 0