An index in SQL Server took one query from 33,517 reads to 66 on the same table. What clustered and nonclustered really mean, why every index is paid for on the way in, and how to find the ones nobody uses.
Using Synonyms to Simplify Cross-Database Queries
Using synonyms can hide a cross-database object name from callers, but permissions, dependency checks, and deployment still need care.
Drawing a Diagram of an Existing Database
A diagram of an existing database starts with real foreign keys, uses SSMS to reveal relationships, and records the important paths a picture misses.
SQL SERVER – Converting Subqueries to Joins
Developers often write subqueries where Joins would do, or the other way around. A short note on converting between the two, with links to my earlier examples.
SQL SERVER – Join Better Performance – LEFT JOIN or NOT IN?
LEFT JOIN or NOT IN: which one gives Join Better Performance? It depends on the data. Compare the two queries and their execution plans on AdventureWorks.
EOMONTH: Month Ends Are Different From DATEADD Month Clipping
EOMONTH gives me a month-end date when that is the actual reporting requirement. DATEADD moves a date by a month. These operations can agree for one step and disagree on the next.
SQL SERVER – Is Cursor Database Object or Datatype?
Whenever we want to loop something we always look for logic like WHILE LOOP or FOR LOOP. Trust me on my word that both of them are cursor when it is about SQL Server.