Re: How do I speed up processing the data from a database link?

"Chuck" <skilover_nospam_at_bluebottle.com> wrote in message news:z3Ecg.7894$Ar6.4014_at_trnddc02...
> Jonathan Lewis wrote:
>>
>> No, "MORE data from" is not an idle wait, it is
>> a wait that means a large data chunk has to come
>> in packets from the client, and therefore is an
>> indication of a possible network "waiting for traffic"
>> issue - the turnaround time should be 'machine speed',
>> not 'human think time'.
>> .
>>
>
> I guess it depends on what you consider an idle wait event to be. The DB
> instance is idle during "MORE data from..." and is not consuming CPU. As
> far as the instance (or more correctly the session) is concerned, it's
> waiting on an asynchronous network event. But that doesn't mean there's
> nothing that can be improved here. Both the network and remote query are
> still in play. That's why I asked to see the query and the explain plan.
>
> Chuck

By that definition of "idle", all wait events are idle events, as the DB is not doing anything. It is unfortunate that the word idle was ever introduced to the discussion of wait events.

I believe the informal understanding of an "idle" event is one that does not matter because there is not task that would complete more quickly in the absence of that event.

Hence

    sql*net message from client is (often) an idle event because     it represents the user sitting back looking at the results on screen.

    sql*net more data from client is not an idle event because the     client software is "trying" to send a large packet of data and     had to break it into bits and (from its perspective) is waiting     for the request from the server for the next instalment.

-- 
Regards

Jonathan Lewis
http://www.oracle.com/technology/community/oracle_ace/ace1.html#lewis

The Co-operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html

Cost Based Oracle: Fundamentals
http://www.jlcomp.demon.co.uk/cbo_book/ind_book.html