Prioritizing Rows In A Union

Daniel Hutmacher shows several ways to combine data from multiple sources, prioritized by source:

For the sake of simplicity, I’ll assume that our example data has a clustered index on the “primary key” that we want to use to determine which rows have already been included in the set – in our example, the primary key is (FirstName, LastName). The data I’m using is a fictional example, but here’s the jist:

  • #Employees has about 33 000 rows.
  • #Customers has about 44 000 rows.
  • #Passengers has about 500 000 rows.

The data is constructed in a way that these queries should return 530 000 rows, so we’re looking at some overlap but far from totally overlapping rows.

Example #492,806 that T-SQL is not a true 4th Generation Language, that how you write the query can greatly matter for performance.

Related Posts

Enabling Database-Level Change Tracking

Tim Weigel continues a series on change tracking: If you don’t provide a retention period, SQL Server’s default is 2 days. Auto-cleanup defaults to ON unless you tell it otherwise. Easy! The table level commands aren’t any more complicated. Before we get started, please note that change tracking requires a primary key on the table […]

Read More

Isolation Levels and Dynamic SQL

Max Vernon points out how transaction isolation levels work when combined with sp_executesql: Imagine you have a piece of code where you don’t care about the downsides to the “read uncommitted” isolation level, and do your due diligence by adding SET TRANSACTION ISOLATION LEVEL READ UNCOMMITTED; at the start of your code. The code following that statement […]

Read More

Categories

July 2017
MTWTFSS
« Jun Aug »
 12
3456789
10111213141516
17181920212223
24252627282930
31