Additional, we now saw the PLE drop from a few thousand to 7 within the 30 second time frame.

Size of the cache at that moment was 8 Gb.

This means that SQL-server expects that within 14 seconds the total memory will be refressed.

the process that was running at that time was running for 13 seconds.

(This was probably the process wich did cause the 'effect').

But how is it possible that SQL-server Expects that the memory can be refressed (8Gb of it) in 14 seconds ?

I do not think normal disk IO (SAN) can handle that amount of IO.

How is the PLE determined?

Thanks for your time and attention.

ben brugman

sorry for shouting.:-D