Re: Performance of Oracle AQ (Advanced Queueing)

Alberto

We've done a lot of performance testing of AQ with the consulting group I'm in with Oracle, and it's possible to get quite high throughput if you play with a few things:

1. Tune the LGWR by using special cache devices like EMC or stripe the redo at a small stripe width (32K) across multiple disks using RAW I/O. Try to isolate I/O to these disks for just the redo logs.
2. Improving database I/O greatly improves AQ performance. Stripe the database (system, rollback, redo, temp) and queue tables across as many physical disk spindles (and controllers!) as possible using RAW I/O and a volume manager.
3. The best way to increase throughput is to collaborate with the business and seek business message key partitioning. If the business can be split by transaction types, for instance orders and sales, then these can be used to drive messages into different queues. The same applies within a business, for instance if orders can be split by geography then this can be used to multiplex messages across recipients. This allows the scalability limitations of a single queue design to be circumvented by using multiple queues.

HTH. Pete

dellera_at_my-deja.com wrote:

> Hi everybody,
>
> has anybody tested/measured the performances of
> Oracle AQ (Advanced Queueing) packages ?
>
> I realize that the performances of AQ
> can't be better than the performances
> of insertion/deletion/select from, since
> there's always an Oracle table underlying
> each queue. So, a certain performance degradation
> is expected, but how much ?
>
> Any help would be greatly appreciated. We are
> now beginning to start some tests about AQ, so
> of course I will send the results to anybody who
> shows interest.
>
> Thanks in advance
> Alberto
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.

Received on Fri Oct 29 1999 - 10:48:57 CDT