[quote]Most often non-RDBMS people use their generated physical locator as the cluster because this is how a magnetic tape or disk access modle of data would have done it.
That may be true for non-RDBMS people but others realize the value of it in preventing page splits of the data involved in the CI. Page splits are horribly expensive things both CPU and I/O wise on heavily inserted OLTP tables. Page splits are a frequent cause of massive GUI timeouts.
If a table is mostly static, then I agree... the clustered index should be used for something else. If the table suffers a lot of inserts, then a CI on a narrow, ever increasing, and unique column such as an IDENTITY column or a DATETIME column and an IDENTITY column as a uniquefier is generally the way to go. And, yes... it's very much like mag tape requirements. It worked for mag tape and it works here.
is pronounced ree-bar and is a Modenism for R
First step towards the paradigm shift of writing Set Based code: Stop thinking about what you want to do to a row... think, instead, of what you want to do to a column.
If you think its expensive to hire a professional to do the job, wait until you hire an amateur. -- Red Adair
How to post code problemsHow to post performance problemsForum FAQs