Re: Paging space on a fastT

From: John Jolet (john.jolet_at_FXFN.COM)
Date: 01/14/04

Next message: JOSEPH KREMBLAS: "Re: Paging space on a fastT"

Previous message: Hunter, Mark: "Re: Paging space on a fastT"
In reply to: Hunter, Mark: "Re: Paging space on a fastT"
Next in thread: jbarratt: "Re: Paging space on a fastT"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Date:         Wed, 14 Jan 2004 11:52:47 -0600
To: aix-l@Princeton.EDU

however, if you pin the sga, that won't happen. if, as we do, you
reserve the entire server for oracle, that's not a problem.

On Wed, 2004-01-14 at 11:33, Hunter, Mark wrote:
> Here is what we saw and why. Note IBM and oracle both recommend setting vmtune
> minperm/maxperm way down on oracle servers.
>
>
> Oracle allocates a huge SGA (shared memory). This is not backed by paging space
> initially. Typically, the Oracle team wants about 60% of the box for
> themselves.
>
> Oracle puts stuff into the SGA. No paging space yet.
>
> Time passes. Files get touched. IBM's caching algorthm + shared memory +
> processes eat up all the real memory. No paging space yet.
>
> More memory is requested. IBM's paging algorithm kicks in. The oracle SGA has
> hot areas, and a bunch of stuff that almost never gets touched. The little used
> areas of this shared memory get paged out to allow room for more file caching.
> Lots of paging space now if minperm/maxperm are too large.
>
> The paging space around the oracle SGA only gets released when the last process
> stops touching it. Thus the reboot/restart clears paging space for a bit.
>
> Note: Once you change min/max perm, you have to cycle the database to get the
> space back since once allocated, paging space normally only gets unallocated
> when the last process accessing it is gone.
>
> Mark Hunter
>
> -----Original Message-----
> From: jbarratt [mailto:jbarratt@COMPSAT.COM]
> Sent: Wednesday, January 14, 2004 11:20 AM
> To: aix-l@Princeton.EDU
> Subject: Re: Paging space on a fastT
>
>
> It seems to me that since you can turn on and off paging spaces on the
> fly, and that deferred paging space algorithms don't take more than
> necessary. that you would just add as needed (that's what I have always
> done)..however, don't come crying to SAP when paging space hits 100% and
> you are unable to open a terminal session to add more...you will
> probably be recycling the power.
>
>
> Just my 2 cents
>
>
> -----Original Message-----
> From: IBM AIX Discussion List [mailto:aix-l@Princeton.EDU] On Behalf Of
> Wilson, Jeff
> Sent: Wednesday, January 14, 2004 12:09 PM
> To: aix-l@Princeton.EDU
> Subject: Re: Paging space on a fastT
>
>
> We are running 5.1 . I am going to try and talk them into changing the
> min and maxperms first and see how that helps. We mostly see the paging
> space get lower and lower over time. We start out with 98% free and
> over a couple of months it gets down to 60-70% free.
>
> Jeff Wilson
> 303-737-5399
>
> -----Original Message-----
> From: Bob.Kelley@BRINKSINC.COM [mailto:Bob.Kelley@BRINKSINC.COM]
> Sent: Wednesday, January 14, 2004 9:52 AM
> To: aix-l@Princeton.EDU
> Subject: Re: Paging space on a fastT
>
> Version of AIX?
>
> Reason for this is with aix 5 you now have whats commonly referred to
> defferred paging. This essentially is the allocation of paging space
> when it's needed, rather than in previous versions of aix 4.3.2 and
> below I believe where exe's would allocate paging space as large as
> their running image even if it turned out they never used it. This is
> one of the reasons Simon points out that its no longer necessary to
> adhere to vendors software requirements for paging space in certain
> cases bc its really solution dependent.
>
> Bobby Kelley Jr.
> 972-877-5341
>
>
>
> |---------+-------------------------------->
> | | "Hunter, Mark" |
> | | <Mark.Hunter@ANHEUSER|
> | | -BUSCH.COM> |
> | | Sent by: IBM AIX |
> | | Discussion List |
> | | <aix-l@Princeton.EDU>|
> | | |
> | | |
> | | 01/14/2004 10:29 AM |
> | | Please respond to IBM|
> | | AIX Discussion List |
> | | |
> |---------+-------------------------------->
>
> >-----------------------------------------------------------------------
> ------------------------------------------------------------------------
> ---|
> |
> |
> | To: aix-l@Princeton.EDU
> |
> | cc:
> |
> | Subject: Re: Paging space on a fastT
> |
>
> >-----------------------------------------------------------------------
> ------------------------------------------------------------------------
> ---|
>
>
>
>
> We run SAP on Oracle.
>
> We set minperm to 5% and maxperm to 20%. Helps alot.
> Our SAP team wants 3x memory for paging also. We ignore this and give
> them 1x. With minperm/maxperm set, they use very little paging space.
>
> Mark Hunter
>
> -----Original Message-----
> From: Bob.Kelley@BRINKSINC.COM [mailto:Bob.Kelley@BRINKSINC.COM]
> Sent: Wednesday, January 14, 2004 10:21 AM
> To: aix-l@Princeton.EDU
> Subject: Re: Paging space on a fastT
>
>
> what version of aix are you running? check your vmtune settings and post
> pls
>
> Bobby Kelley Jr.
> 972-877-5341
>
>
>
> |---------+---------------------------->
> | | "Wilson, Jeff" |
> | | <Jeff.Wilson@GWL.|
> | | COM> |
> | | Sent by: IBM AIX |
> | | Discussion List |
> | | <aix-l@Princeton.|
> | | EDU> |
> | | |
> | | |
> | | 01/14/2004 09:59 |
> | | AM |
> | | Please respond to|
> | | IBM AIX |
> | | Discussion List |
> | | |
> |---------+---------------------------->
>
> >
> ------------------------------------------------------------------------
> -------
>
> -------------------------------------------------------------------|
> |
> |
> | To: aix-l@Princeton.EDU
> |
> | cc:
> |
> | Subject: Re: Paging space on a fastT
> |
>
> >
> ------------------------------------------------------------------------
> -------
>
> -------------------------------------------------------------------|
>
>
>
>
> We are running the paging monster app "SAP on DB2". I have never seen
> an app eat paging space like this one. If we bounce SAP we get it all
> back but until we do that SAP keeps all the paging space and dose not
> release it. As always the recommend 2 X physical memory but with the
> new SAP installs they want 3 X physical memory which is ridicules but I
> have no choice in the matter so I have to add it even if they don't need
> it.
>
> Jeff Wilson
> 303-737-5399
> -----Original Message-----
> From: JOSEPH KREMBLAS [mailto:jkremblas@REDHEARTGIFTS.COM]
> Sent: Wednesday, January 14, 2004 8:34 AM
> To: aix-l@Princeton.EDU
> Subject: Re: Paging space on a fastT
>
> Jeff,
>
> Before throwing more paging space at the system, have you ran
> vmstat and iostat to see what the system is doing? Can you provide
> a
> vmstat listing and iostat listing? Are you running Oracle? Any
> other
> databases? Also, provide the output of vmtune and schedtune. The
> group may be able to tell what's going on performance wise and
> make
> some recommedations.
>
> --joseph
> -----Original Message-----
> From: IBM AIX Discussion List [mailto:aix-l@Princeton.EDU] On
> Behalf
> Of Wilson, Jeff
> Sent: Wednesday, January 14, 2004 8:27 AM
> To: aix-l@Princeton.EDU
> Subject: Paging space on a fastT
> I was wondering if anyone has had any experience putting paging
> space
> on attached disks like a fastT 900. I am out of space on my
> internal
> OS disks and need more paging space and was wondering if there are
> any problems I should be aware of before I do this. Thanks for
> you
> help.
>
>
>
> Jeff Wilson
> System Administrator
> Great-West Life & Annuity Ins. Co.
> Desk: (303) 737-5399
>
>
>
> NOTICE TO BRINK'S EMPLOYEES:
> All electronic mail communications composed, sent, received or stored on
> or through Brink's equipment are subject to the Brink's Electronic
> Communications Policy. Employees should have no expectation of privacy
> in their use of the Company's electronic mail system
>
>
>
> NOTICE TO BRINK'S EMPLOYEES:
> All electronic mail communications composed, sent, received or stored on
> or through Brink's equipment are subject to the Brink's Electronic
> Communications Policy. Employees should have no expectation of privacy
> in their use of the Company's electronic mail system.

Next message: JOSEPH KREMBLAS: "Re: Paging space on a fastT"

Previous message: Hunter, Mark: "Re: Paging space on a fastT"
In reply to: Hunter, Mark: "Re: Paging space on a fastT"
Next in thread: jbarratt: "Re: Paging space on a fastT"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Relevant Pages

Re: Does my server have enough memory
... I have root access and check the paging space at the top of every hour ... What should I watch to determine if we're short on memory 2. ... Oracle and some web services for user access. ...
(comp.unix.aix)
Re: strange memory leak
... server that is going to host a new JDE EnterpriseOne implementation. ... So far, oracle & JDE are installed, and users are able to log in, ... the machine shipped with 8GB memory, ... When paging space fills up, it means that the system sends out ...
(comp.unix.aix)
Re: Paging space on a fastT
... Note IBM and oracle both recommend setting vmtune ... Oracle allocates a huge SGA (shared memory). ... No paging space yet. ...
(AIX-L)
Re: Volume group must be varied on: Strange never seen this before
... I've a p570 AIX 5.3 ML02 with same symptom, using SAP an Oracle too. ... and I was going to add more paging space when I noticed ... "Volume group must be varied on; use varyonvg command." ...
(AIX-L)
Re: Paging space utilized... Why?
... My question is, given the configuration, how is ANY shared memory being kicked ... maxperm and maxclient should be around 10 and minperm should be ... paging space used, which appears to be my Oracle sga area. ...
(AIX-L)