| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
 |
 | |||
| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
|
| |||
Home -> Community -> Usenet -> c.d.o.server -> Re: CHECKPOINTS
Hi Craig,
Here is proof of the timing error ...
### start craig.lst ###
SQL> select time_waited from v$system_event where event = 'PL/SQL lock timer';
no rows selected
SQL> set timing on
SQL> execute sys.dbms_lock.sleep(10);
PL/SQL procedure successfully completed.
real: 10245
SQL> select time_waited from v$system_event where event = 'PL/SQL lock timer';
EVENT TIME_WAITED ---------------------------------------------------------------- ----------- PL/SQL lock timer 1024### end craig.lst ###
Also, have a look at the table on page 14 where you show the time_waited going up by 100 for each second in the seconds_in_wait column.
The unit of time is important for a lot of reasons. Two significant ones are that it would cause you to understate waiting time relative to wall clock / elapsed time by a factor of ten. Similarly, it may cause you to overlook an average wait of 15.3 for 'db file sequential read' as within the disk manufacturers specifications for random reads, whereas in fact this represents a severe I/O problem.
The next most important error is that you suggest that 'buffer busy waits' have something to do with block cloning. That could lead people down some totally unproductive blind alleys. Other than that, the concept of a wait timeout is wrongly explained, and there are numerous typographical errors in the names of the events, and other things. Prehaps, more significant is the omission of any mention of CPU time, and the claim that a session is always waiting.
I'll leave it at that.
Regards,
Steve Adams
http://www.ixora.com.au/ http://www.oreilly.com/catalog/orinternals/ http://www.christianity.net.au/
On Wed, 01 Mar 2000 21:19:05 -0800, Craig Shallahamer <craig_at_orapub.com> wrote:
>
>--------------ABAAA3DF8ADC082B8866669A
>Content-Type: text/plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
>Steve, Bad hair day huh...
>
>[snip]
>
>My commercial: If your proposed "timing error" is the most serious error I
>made, then thank you for complementing my paper. I'm taking my information
>based upon information I have received from a number of sources. I don't
>always believe Oracle documentation, but you could be right because I've
>never seen any experimental data which indicates the actual time unit one way
>or another. But regardless, unless one is using the timing information for
>non-relative analysis (which is unusual) it doesn't make any difference.
>What one is usually interested in is which wait events take the most time per
>a time slice (e.g., per second, per 15 minutes) and which wait events get
>"pinged" the most, not the actual wait time.
>
>[snip]
>
>Best regards,
>
>Craig.
>
>
>Steve Adams wrote:
>
>> Hello Craig,
>>
>> [snip]
>>
>> Your commercial: That paper has a number of serious errors. Perhaps the
>> most significant is that you claim that the unit of time is milliseconds
>> when in fact is the centiseconds.
>>
>> Regards,
>> Steve Adams
>> http://www.ixora.com.au/
>> http://www.oreilly.com/catalog/orinternals/
>> http://www.christianity.net.au/
>>
>> On Wed, 01 Mar 2000 15:56:11 -0800, Craig Shallahamer <craig_at_orapub.com>
>> wrote:
>>
>> >[snip]
>> >
>> >The commercial: There are more details about the session wait stuff in a
>> >paper, Quick Contention Identification Using Oracle's Session Wait Views
>> >on my web site. I also have an Internet Video Seminar on this topic as
>> >well. My performance classes also cover this in detail. And, I have
>> >some tools, for free on my web-site, that collect and report on session
>> >wait details.
>> >
>> >Hope this helps,
>> >
>> >Craig.
>> >
>> >Craig A. Shallahamer
>> >President, OraPub, Inc.
>> >
>> >craig_at_orapub.com
>> >http://www.orapub.com
>> >+1.503.636.0228 (tel)
>> >+1.413.812.4406 (fax)
Received on Thu Mar 02 2000 - 00:00:00 CST
 |
 |