Vsevolod Afanassiev wrote:
> On Oct 12, 1:17 am, DA Morgan <damor..._at_psoug.org> wrote:
>> Vsevolod Afanassiev wrote:
>>> On Oct 10, 11:02 pm, DA Morgan <damor..._at_psoug.org> wrote:
>>>> Vsevolod Afanassiev wrote:
>>>>> I am trying to get Client Result Cache to work in Oracle 11g database.
>>>>> This feature is described in OCI Programmer's Guide, pages 10-22 -
>>>>> 10-27.
>>>>> 1. Parameter CLIENT_RESULT_CACHE_SIZE has been set to 1000000 and
>>>>> database restarted. I can see new value in V$PARAMETER.
>>>>> 2. Parameter RESULT_CACHE_MODE has been set to FORCE.
>>>>> I am definitely getting results cached on the server: I can see it in V
>>>>> $RESULT_CACHE_* views and performance of large table scans is sub-
>>>>> second. However, view CLIENT_RESULT_CACHE_STATS$ is empty, so I assume
>>>>> that
>>>>> results do not get cached on the client. Any idea what is missing?
>>>>> How the teste were done:
>>>>> - Using SQL*Plus running on the same server as database. I tried
>>>>> Bequeath
>>>>> connections and connections though listener (listener is also 11g).
>>>>> - Large table is created (no indexes)
>>>>> - Full table scan is performed using SELECT COUNT(*)
>>>>> so result set consists of one row.
>>>>> - There is no sqlnet.ora
>>>> Run the RESULT CACHE demos in Morgan's Library atwww.psoug.org.
>>>> They work.
>>>> --
>>>> Daniel A. Morgan
>>>> University of Washington
>>>> damor..._at_x.washington.edu (replace x with u to respond)
>>>> Puget Sound Oracle Users Groupwww.psoug.org-Hide quoted text -
>>>> - Show quoted text -
>>> Daniel,
>>> Thanks
>>> My question was about CLIENT-side result cache, not server-side cache.
>>> Please correct me if I am wrong:
>>> - Views V$RESULT_CACHE_* describe server-side cache
>>> - View CLIENT_RESULT_CACHE_STATS$ describes client-side cache
>>> - If this view is empty then client-side cache isn't working
>> Incorrect. There is only one result cache and it resides on the server
>> so that it can be shared by all sessions. CLIENT_RESULT_CACHE_STATS is
>> exactly what the names implies. It is statistics related to the client's
>> use of that cache.
>> --
>> Daniel A. Morgan
>> University of Washington
>> damor..._at_x.washington.edu (replace x with u to respond)
>> Puget Sound Oracle Users Groupwww.psoug.org- Hide quoted text -
>>
>> - Show quoted text -
>
> Daniel,
>
> Let me quote from Oracle Call Interface (OCI) Programmer's Guide, page
> 10-24
> "When client result caching is enabled, the query result can be cached
> on the client or on the
> server or both. The client result caching can be enabled even if the
> server result cache (that is
> enabled by default) is disabled".
>
> There are other statements in this guide that clearly imply that there
> are two caches: one
> on the server and another on the client. See pages 10-22 - 10-27.
>
> Of course it applies only to OCI client. Non-OCI clients (e.g. thin
> JDBC) don't have client-side cache.
Apparently OCI is giving you additional capabilities I've not seen
before. Something new to look at. Thanks.
--
Daniel A. Morgan
University of Washington
damorgan_at_x.washington.edu (replace x with u to respond)
Puget Sound Oracle Users Group
www.psoug.org
Received on Thu Oct 11 2007 - 17:52:52 CDT
