Database Administrators Stack Exchange is a question and answer site for database professionals who wish to improve their database skills and learn from others in the community. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

We are using SQL Azure Premium P1 which comes with 8GB ram and 150 IOPS. For a couple of hours, we have been seeing timeouts on SQL queries (even fairly simple queries take time).

In logs for SQL Azure we have seen that our physical read IOPS has increased from a steady 5 IOPS to around 130 in a matter of minutes but our actual production in the system relying on SQL Azure has not increased.

The minute it jumped to 130 IOPS the log shows that the database memory usage decreased from a very steady 4.8GB to around 1GB

I think it makes sense that If memory usage drops, physical reads increase.

I have looked at queries being executed on the server but see nothing out of the ordinary except for the slower response.

My question is whether there any reasons why SQL Azure (or SQL server for that matter) would drop things from memory and favor disk IO, even if there is plenty memory available.

share|improve this question
up vote 1 down vote accepted

Turns out that even if you pay for reserved resurces and are guaranteed 8GB of RAM, you aren't really guaranteed it. And it's not really reserved for you. In our case it turned out that another Windows Azure customer was using too much resources and our database got 1GB instead of 8.

share|improve this answer
We're investigating P1/P2 servers as well and would love to know how you came about this information about not really being guaranteed the RAM they state. I hesitate to use the service if I'm not really getting what I'm paying for. – Vyrotek Nov 25 '13 at 23:23

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.