Re: Virtual drive on Solaris

Ferenc,

What a small world!

That class in Dallas was the very first teach of the excellent "Oracle Server Internals" seminars anywhere, ever. Originally, Scott Gossett was supposed to teach all of them, but he was booked at the time so Oracle Education hired me as a replacement contractor to deliver them. Then, Scott suffered some serious health problems, so I got to teach that course about a dozen more times in North and South America through 1998 and 1999. But Scott got well and has been doing it (extremely well!) ever since, I'm told...

I have to admit that I was flat-out terrified teaching that first class in Dallas. Not only was it brand-new curriculum, but there were some 30-35 people packed in there and most of them were like you -- exceedingly sharp! One guy (do you remember?) who sat in the front row had been on the team at Bell Labs who worked on the original UNIX implementation; he also consulted to Oracle to build the newly revamped transaction layer for Oracle7 in the late 1980s. It was a 3-day class and he casually revealed his pedigree midway through the first day. I spent the next 2.5 days watching his reaction to everything I said out of the corner of my eye. If I said something and he nodded, the angels sang hosannas and the sun shone brightly. If I said something and he frowned or squinted and tilted his head, I almost fell over. Absolutely nerve-wracking! Oh well, that which doesn't kill you...

Thanks for the info!

-Tim

• Original Message ----- To: "Multiple recipients of list ORACLE-L" <ORACLE-L_at_fatcity.com> Sent: Tuesday, June 25, 2002 6:28 AM

> Used it for large application upgrades, works a charm, cut down the time
by
> about a third. But it does not work on every platform, because when I set
it
> and then perform a transaction, upon dumping the redo log, I find my
> transaction in there. But thanks Tim.
>
> BTW, I was in that class, you ran it in Dallas in 1998 at Oracle
Education.
> Cheers :
>
> Regards:
> Ferenc Mantfeld
> Senior Performance Engineer
> Siebel Performance Engineering
> Melbourne, 3000, VIC, Australia
> Only Robinson Crusoe had all his work done by Friday
>
>
> -----Original Message-----
> Sent: Monday, 24 June 2002 4:23 PM
> To: Multiple recipients of list ORACLE-L
>
>
> I hesitated mentioning that parameter in this forum, but I figured what
the
> heck? Could be fun, in a sick way... :-)
>
> Once I was teaching a DBA class and mentioned "_DISABLE_LOGGING".
> Immediately, I saw every head in the class look down, scribbling
furiously!
> I had to backtrack very quickly and warn of the consequences of disabling
> redo logging (i.e. database corruption if not shutdown normally for any
> reason)...
>
> ----- Original Message -----
> To: "Multiple recipients of list ORACLE-L" <ORACLE-L_at_fatcity.com>
> Sent: Monday, June 24, 2002 2:48 PM
>
>
> > Hi Tim
> >
> > Yes, I have tried the _disable_logging, does not work on all platforms.
DB
> > starts up fine, but redo log is generated, evidenced by log switching
> going
> > on.
> >
> > Also if I do a normal DML (large-ish one to verify), then dump the redo
> log,
> > I see my transaction there, so for a 420R, running Solaris8 and Oracle
> > 9.0.1, it would seem that _disable_logging does not work.
> >
> > I don't want to complicate the picture even further with transportable
> > tablespaces, which would mean that I would need to store all dependent
> > objects (in this case indexes only) in the same tablespace, which I
could
> > easily achieve by rebuilding all indexes using a dynamic SQL.
> >
> > Informatica BTW does not only do single level inserts, version 5.0
onwards
> > has a 'bulk load' feature, but I am not sure what this actually does.
> > Previously Sagent also had a 'direct load' switch, which meant that it
> wrote
> > all of its data to large (very large) flat files and then used
Sql*Loader
> > direct path to load. Fast, but Sagent at the time was very unreliable,
> > because on identical runs, it would sometimes load all the data,
sometimes
> > only a portion, and every time, would report no errors and everything
> hunky
> > dory, until you went looking for your data. I remember that took me
about
> a
> > week of arguing to prove that Sagent was at fault.
> >
> > Thanks for the suggestion of the Non volatile RAM (NVRAM) unit, it makes
> the
> > most sense. I will suggest this to my damagers.
> >
> > Regards:
> > Ferenc Mantfeld
> > Senior Performance Engineer
> > Siebel Performance Engineering
> > Melbourne, 3000, VIC, Australia
> > Only Robinson Crusoe had all his work done by Friday
> >
> >
> > -----Original Message-----
> > Sent: Sunday, 23 June 2002 9:03 PM
> > To: Multiple recipients of list ORACLE-L
> >
> >
> > Have you considered setting "_DISABLE_LOGGING = TRUE"
> > instead? It could be just as disastrous... ;-)
> >
> > Buying an NVRAM unit would probably be more sensible, since
> > at least then you have some probability of the file-system
> > on such a unit surviving node failure or restart.
> >
> > I don't use Informatica, but I believe it mainly does
> > single-row inserts, so not using the APPEND hint is a
> > blessing anyway. After all, who likes one row in each
> > database block? However, I could be wrong about that and it
> > may actually be performing multi-row/array insertions...
> >
> > I don't know what your loads are like, but how about
> > something like this instead?
> >
> > - create a small database with _DISABLE_LOGGING set to
> > TRUE
> > - use Informatica to load into a tablespace on that small,
> > sacrificial db
> > - use "transportable tablespace" to copy the tablespace to
> > your real DW
> >
> > Just an idea (better you than me to try it!)...
> >
> > ----- Original Message -----
> > To: "Multiple recipients of list ORACLE-L"
> > <ORACLE-L_at_fatcity.com>
> > Sent: Sunday, June 23, 2002 8:53 PM
> >
> >
> > > Hi All
> > >
> > > does anyone have any white paper or info on how to
> > configure a dedicated
> > > portion of real memory as a virtual drive on Solaris ? I
> > want to move my
> > > online redo logs (4 X 128 M single threaded) for a 300 GB
> > DW onto it, to
> > > speed up Informatica ETL, since Informatica does not allow
> > me to specify /*+
> > > APPEND */ mode of insert. I know I will not bypass the SQL
> > layer this way,
> > > but at least, the LGWR will be writing to memory instead
> > of disk. Thanks in
> > > advance.
> > >
> > > Regards:
> > > Ferenc Mantfeld
> > > Senior Performance Engineer
> > > Siebel Performance Engineering
> > > Melbourne, 3000, VIC, Australia
> > > Only Robinson Crusoe had all his work done by Friday
> > >
> > >
> > > -----Original Message-----
> > > Sent: Saturday, 22 June 2002 9:03 PM
> > > To: Multiple recipients of list ORACLE-L
> > >
> > >
> > > On Solaris
> > >
> > > ps -ef -opid,ppid,vsz=VIRTMEM -orss=PHYSMEM
> > -opmem,pcpu,user,args
> > >
> > > use:
> > >
> > > psrinfo -v
> > > prtconf | grep Mem
> > > format
> > > uname -a
> > >
> > > HTH
> > >
> > > Richard
> > >
> > > -----Original Message-----
> > > Sent: Saturday, June 22, 2002 1:38 PM
> > > To: Multiple recipients of list ORACLE-L
> > >
> > >
> > > Good day to everyone...
> > >
> > > I have two questions related to Linux and Solaris...
> > >
> > > * I need do find memory usage (physical, virtual...) of a
> > particular
> > > proccess. PID is given by by "ps", but what aditional
> > parameters I have to
> > > provide? At a first glance, output of "man ps" vas
> > confusing...
> > >
> > > * How do I find computer's configuration - what CPU,
> > numbers of CPUs, clock,
> > > amount of memory, number of harddrives, what version of
> > OS, what OS patches
> > > are applied...?
> > >
> > > No, I can't ask sysadmin about that (hard to explain),
> > and, no, I don't have
> > > any kind of advanced manuals...
> > >
> > > Thanks in advance,
> > > Vladimir
> > >
> > > --
> > > Please see the official ORACLE-L FAQ:
> > http://www.orafaq.com
> > > --
> > > Author: Vladimir Barac - posao
> > > INET: vladob_at_aster.si
> > >
> > > Fat City Network Services -- (858) 538-5051 FAX: (858)
> > 538-5051
> > > San Diego, California -- Public Internet access /
> > Mailing Lists
> > >
> > ------------------------------------------------------------
> > --------
> > > To REMOVE yourself from this mailing list, send an E-Mail
> > message
> > > to: ListGuru_at_fatcity.com (note EXACT spelling of
> > 'ListGuru') and in
> > > the message BODY, include a line containing: UNSUB
> > ORACLE-L
> > > (or the name of mailing list you want to be removed from).
> > You may
> > > also send the HELP command for other information (like
> > subscribing).
> > > --
> > > Please see the official ORACLE-L FAQ:
> > http://www.orafaq.com
> > > --
> > > Author: Ji, Richard
> > > INET: Richard.Ji_at_MobileSpring.com
> > >
> > > Fat City Network Services -- (858) 538-5051 FAX: (858)
> > 538-5051
> > > San Diego, California -- Public Internet access /
> > Mailing Lists
> > >
> > ------------------------------------------------------------
> > --------
> > > To REMOVE yourself from this mailing list, send an E-Mail
> > message
> > > to: ListGuru_at_fatcity.com (note EXACT spelling of
> > 'ListGuru') and in
> > > the message BODY, include a line containing: UNSUB
> > ORACLE-L
> > > (or the name of mailing list you want to be removed from).
> > You may
> > > also send the HELP command for other information (like
> > subscribing).
> > >
> > >
> > > --
> > > Please see the official ORACLE-L FAQ:
> > http://www.orafaq.com
> > > --
> > > Author: Ferenc Mantfeld
> > > INET: fmantfeld_at_siebel.com
> > >
> > > Fat City Network Services -- (858) 538-5051 FAX: (858)
> > 538-5051
> > > San Diego, California -- Public Internet access /
> > Mailing Lists
> > >
> > ------------------------------------------------------------
> > --------
> > > To REMOVE yourself from this mailing list, send an E-Mail
> > message
> > > to: ListGuru_at_fatcity.com (note EXACT spelling of
> > 'ListGuru') and in
> > > the message BODY, include a line containing: UNSUB
> > ORACLE-L
> > > (or the name of mailing list you want to be removed from).
> > You may
> > > also send the HELP command for other information (like
> > subscribing).
> > --
> > Please see the official ORACLE-L FAQ: http://www.orafaq.com
> > --
> > Author:
> > INET: tim_at_sagelogix.com
> >
> > Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
> > San Diego, California -- Public Internet access / Mailing Lists
> > --------------------------------------------------------------------
> > To REMOVE yourself from this mailing list, send an E-Mail message
> > to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
> > the message BODY, include a line containing: UNSUB ORACLE-L
> > (or the name of mailing list you want to be removed from). You may
> > also send the HELP command for other information (like subscribing).
> >
> >
> > --
> > Please see the official ORACLE-L FAQ: http://www.orafaq.com
> > --
> > Author: Ferenc Mantfeld
> > INET: fmantfeld_at_siebel.com
> >
> > Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
> > San Diego, California -- Public Internet access / Mailing Lists
> > --------------------------------------------------------------------
> > To REMOVE yourself from this mailing list, send an E-Mail message
> > to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
> > the message BODY, include a line containing: UNSUB ORACLE-L
> > (or the name of mailing list you want to be removed from). You may
> > also send the HELP command for other information (like subscribing).
>
> --
> Please see the official ORACLE-L FAQ: http://www.orafaq.com
> --
> Author: Tim Gorman
> INET: Tim_at_SageLogix.com
>
> Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
> San Diego, California -- Public Internet access / Mailing Lists
> --------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
> the message BODY, include a line containing: UNSUB ORACLE-L
> (or the name of mailing list you want to be removed from). You may
> also send the HELP command for other information (like subscribing).
>
>
> --
> Please see the official ORACLE-L FAQ: http://www.orafaq.com
> --
> Author: Ferenc Mantfeld
> INET: fmantfeld_at_siebel.com
>
> Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
> San Diego, California -- Public Internet access / Mailing Lists
> --------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
> the message BODY, include a line containing: UNSUB ORACLE-L
> (or the name of mailing list you want to be removed from). You may
> also send the HELP command for other information (like subscribing).

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Tim Gorman
  INET: Tim_at_SageLogix.com

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).