RE: Early 11g Advanced Table Compression #'s

  • From: "Powell, Mark D" <mark.powell@xxxxxxx>
  • To: "oracle-l" <oracle-l@xxxxxxxxxxxxx>
  • Date: Fri, 17 Aug 2007 13:01:23 -0400

We have an OLTP and insert select * where is often used.  Data
quantities of 10M are not uncommon for the size of the transaction data
though most of our operations work on just a few rows at a time.  Then
again we do have a significant batch cycle where entire tables get
processed usually using loops that commit after every row or every few
rows.  In many cases the commits are necessary because the same rows we
are updating are needed in other concurrent transactions and in some
cases the commit is just left over from the days when the size of the
rollback segments needed to support a transaction had to be kept small
(application over 12 years old now). 
I think good set of tests would run the gambit from single row
operations, small set operations, to large set operations, and to row by
row processing of the entire table because all of this can exist in one
real-world system.

-- Mark D Powell -- 
Phone (313) 592-5148 



        From: oracle-l-bounce@xxxxxxxxxxxxx
[mailto:oracle-l-bounce@xxxxxxxxxxxxx] On Behalf Of Ghassan Salem
        Sent: Friday, August 17, 2007 10:47 AM
        To: rxsherm@xxxxxxxxxxxxx
        Cc: oracle-l
        Subject: Re: Early 11g Advanced Table Compression #'s
        what do you mean by 'bulk insert' and 'bulk update'?
        To simulate an OLTP, you can you a loop inserting a row and
commiting each time, as OLTP means simple small transactions.
        On 8/17/07, Roby Sherman <rxsherm@xxxxxxxxxxxxx> wrote: 

                Seems my mailer has cut off the apostrophe from the
URL... It should be: 


                On Aug 17, 2007, at 7:14 AM, Roby Sherman wrote:

                        For anyone interested, I ran some very quick
benchmarks on 11g's new Advanced Compression table option COMPRESS FOR
ALL OPERATIONS that Oracle is claiming was "specifically written to
create only the most 'minimal' of performance impacts in OLTP

                        The results are here:

                        I guess their definition of minimal and my
definition of minimal must be different... Anyway, if you're interested
in this feature, feel free to take a quick look!


Other related posts: