A look into clustering to detect outliers in R. An extension on univariate statistical tests to include multivariate data.
TSQL Code must work properly and efficiently. That's not enough though. Unless you are working alone, have perfect memory and plan to never change job, then you need to comment and document your code, it must be inherently readable, well laid out, use informative and obvious names, and it must be robust and resilient; written defensively. It must not rely on deprecated features of SQL Server, or assume particular database settings. Robert Sheldon starts a series of articles that explains the basics.
A holiday in the US has Steve Jones thinking about compromise and communication.
In this post, Tim Smith looks at the different options you can use to audit your SQL Server extracts and loads during the ETL process.
You will see here a way to handle history tables. This way only takes into account Date-based data cleanup but is easily generalizable.
This week Steve asks you to make sure you practice your restore skills periodically.
Steve Jones saw a database design test for developers, but he's never been given one.
When mirroring was first released for Azure SQL Database, it used Change Data Capture...
By Steve Jones
One of the things I’ve tried hard to do in database development situations if...
By DataOnWheels
The T-SQL Tuesday topic this month comes James Serra. What career risks have you...
We have two "identical" instances of an ASP.NET web service (or so I have...
Comments posted to this topic are about the item OPENQUERY Flexibility
Comments posted to this topic are about the item A Full Shutdown
Which of these are valid OPENQUERY() uses?
See possible answers