EscVector wrote:
> DA Morgan wrote:
>> EscVector wrote:
>>> I have a multiple part process and I'm able to predict runtime based on
>>> record count within +-30 minutes. The prediction is made by taking the
>>> total to-be-processed record count from the longest single process and
>>> dividing it by the runtime total of all processes.
>>> How is this mathematically possible/sound? This isn't even a linear
>>> projection. Just dumb luck?
>>>
>>> I'm looking for ways to project performance based on record counts, so
>>> any suggestions are welcome.
>> dbms_application_info.set_session_longops
>>
>> http://www.psoug.org/reference/dbms_applic_info.html
>> --
>> 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
>
> What if the majority of work is in ETL tool and very little database
> throughput is required.
> Example: start session, direct path read big table for a few minutes,
> do 15hrs etl stuff, 99% of which is not using db resources. Maybe
> direct this to a math group?
Sounds like a modification of the ETL tool's source code is in order.
--
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 Fri Dec 22 2006 - 12:28:36 CST
