Re: RES: Priority to a mission critical transaction

Ron Mayer wrote:

>Before asking them to remove it, are we sure priority inversion
>is really a problem?
>
>I thought this paper: http://www.cs.cmu.edu/~bianca/icde04.pdf
>did a pretty good job at studying priority inversion on RDBMs's
>including PostgreSQL on various workloads (TCP-W and TCP-C) and
>found that the benefits of setting priorities vastly outweighed
>the penalties of priority inversion across all the databases and
>all the workloads they tested.
>
>
>
I have the same question.  I've done some embedded real-time
programming, so my innate reaction to priority inversions is that
they're evil.  But, especially given priority inheritance, is there any
situation where priority inversion provides *worse* performance than
running everything at the same priority?  I can easily come up with
situations where it devolves to that case- where all processes get
promoted to the same high priority.  But I can't think of one where
using priorities makes things worse, and I can think of plenty where it
makes things better.

Brian