Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> RE: Log writer

RE: Log writer

From: Mohan, Ross <MohanR_at_STARS-SMI.com>
Date: Thu, 22 Mar 2001 09:30:37 -0800
Message-ID: <F001.002D4FAD.20010322092038@fatcity.com>

Kev,

 
Pure
shot from hip:
<FONT face=Arial color=#0000ff

size=2> 
I'd
say you have more network/app response tuning than log tuning. I'd
say
you have more disk tuning ( for table/index reads) than log tuning. I

would
DEFINITELY turn the silly ass MTS off.
<FONT face=Arial color=#0000ff

size=2> 
I'd
find out how to make sure disk cache is 4 MB and is writeBACK
not
writeTHRU. ( if either ) <FONT
face=Arial color=#0000ff size=2>I'd have questions about the controller, too, but.."later".
<FONT face=Arial color=#0000ff

size=2> 
(You
didn't tell me the size of the redo log buffer...)
<FONT face=Arial color=#0000ff

size=2> 
Of
course, having a two disk wonder (or whatever it is) is the number
one
thing to work on.
<FONT face=Arial color=#0000ff

size=2> 
In any
case, with your hardware, your system should easily be able to
handle a few users, in its sleep.
<FONT face=Arial color=#0000ff

size=2> 
<FONT face=Arial color=#0000ff

size=2>There's about two dozen little tweaks you can do to perk things up a bit,

but
basically, your "rock and a hard place" is going to be a two disk wonder
on a
single controller...
<FONT face=Arial color=#0000ff

size=2> 
<FONT face=Arial color=#0000ff

size=2>hth
<FONT face=Arial color=#0000ff

size=2> 
<FONT face=Arial color=#0000ff

size=2>Ross
<FONT face=Arial color=#0000ff

size=2> 
<FONT face=Arial color=#0000ff

size=2> 

  <FONT face="Times New Roman"
  size=2>-----Original Message-----From: Kevin Kostyszyn   [mailto:kevin_at_dulcian.com]Sent: Thursday, March 22, 2001 11:31   AMTo: Multiple recipients of list ORACLE-LSubject: RE:   Log writer
  <FONT face=Arial color=#0000ff
  size=2>Jeez, Seagate Cheeta 18 and 36, 10k RPM     ahhh
  UNFORMATTED CAPACITY (GB)

  ________________22.54       FORMATTED 
  CAPACITY (GB) __________________18.21       
  AVERAGE SECTORS PER TRACK ________________213 rounded 
  down       ACTUATOR TYPE
  ____________________________ROTARY VOICE 
  COIL       TRACKS
  ___________________________________167,008       
  CYLINDERS ________________________________6,962 
  user       HEADS
  ______PHYSICAL______________________24       
  DISCS (3 in)
  _____________________________12       MEDIA 
  TYPE _______________________________THIN 
  FILM/MR       RECORDING METHOD
  _________________________PRML 8/9 PR4       
  INTERNAL TRANSFER RATE (mbits/sec)________152 to   231       EXTERNAL TRANSFER RATE (mbyte/sec)   _______40
  Sync            Low
  Voltage Differential(LVD) _______80
  Sync       SPINDLE SPEED (RPM)
  ______________________10,025       AVERAGE 
  LATENCY (mSEC) ___________________2.99       
  BUFFER (/optional)
  _______________________1MB/4MB         
  Read Look-Ahead, Adaptive,           

  I believe the only problem is
  that they are on one controller.  However, this is not a huge database,   only about 3-4 users connected at a time, but lots of DML.  The machine   is a Dell Precision 410 with a gig of Ram and dual PIII 450's.  I don't   think that it's a huge problem, but it does keep coming up.  There are   three log file groups of 32 MB's each and yes the database is also spread   accross these two platters. 
   
  Oh and the
  results......YOINK!
   
  <FONT face=Arial color=#0000ff
  size=2>rdbms ipc

  message                                                   
  14980809SQL*Net message from
  client                                          
  9595869pmon
  timer                                                           
  1880177virtual circuit
  status                                               
  1877869smon
  timer                                                           
  1875449dispatcher
  timer                                                     
  1874318SQL*Net break/reset to
  client                                          
  49746control file parallel
  write                                            
  18026db file sequential
  read                                                 
  3947log file
  sync                                                           
  1456db file scattered
  read                                                  
  1017
   
  <FONT face=Arial color=#0000ff
  size=2>EVENT                                                            
  TIME_WAITED---------------------------------------------------------------- 
  -----------log file parallel
  write                                                  
  616Null
  event                                                               
  411library cache
  pin                                                        
  315refresh controlfile
  command                                              
  292reliable
  message                                                         
  273rdbms ipc
  reply                                                          
  146control file sequential
  read                                             
  108log file switch
  completion                                                
  58SQL*Net more data to
  client                                               
  45enqueue                                                                   
  27file
  identify                                                             
  24
   
  <FONT face=Arial color=#0000ff
  size=2>EVENT                                                            
  TIME_WAITED---------------------------------------------------------------- 
  -----------log file single
  write                                                     
  23latch
  free                                                                
  19SQL*Net message to
  client                                                 
  17file
  open                                                                 
  16buffer busy
  waits                                                         
  13SQL*Net more data from
  client                                              
  6direct path
  read                                                           
  5log file sequential
  read                                                   
  0db file parallel
  write                                                     
  0instance state
  change                                                      
  0direct path
  write                                                          
  0
   
  <FONT face=Arial color=#0000ff
  size=2>EVENT                                                            
  TIME_WAITED---------------------------------------------------------------- 
  -----------LGWR wait for redo
  copy                                                    
  0
  <FONT face=Arial color=#0000ff
  size=2> 
  <FONT face=Arial color=#0000ff
  size=2>Which ya think master Ross?
  <FONT face=Arial color=#0000ff
  size=2> 
  <FONT face=Arial color=#0000ff
  size=2>Kev
  <FONT
  size=2><FONT face=Arial
  color=#0000ff> 
  <SPAN
  class=562192116-22032001> 
  <SPAN
  class=562192116-22032001> -----Original   Message-----From: root_at_fatcity.com [mailto:root_at_fatcity.com]On   Behalf Of Mohan, RossSent: Thursday, March 22, 2001 11:02   AMTo: Multiple recipients of list ORACLE-LSubject: RE:   Log writer   

    Disk mfr? rpm? <FONT
    size=2>seek/access times? diskperf -y set     already? CPU,I/O, or diskbound system?     log buffer size? log file size? <FONT     size=2>what else is on the "separate" disks? anything? <FONT     size=2>separate disks share a controller? select     event, time_waited from v$system_event order by 2 desc; results?     just a few questions that come to mind....     -----Original Message----- From:
    Kevin Kostyszyn [<A
    href="mailto:kevin_at_dulcian.com">mailto:kevin_at_dulcian.com]     Sent: Thursday, March 22, 2001 8:50 AM <FONT     size=2>To: Multiple recipients of list ORACLE-L <FONT     size=2>Subject: Log writer
    Hi all,
            I am testing out
    Spotlight on Oracle from Quest Software.  It is telling me     that my log writer process is slow.  What it looks     like it is saying is that most disks take about 20ms     to right the data and mine is taking on average <FONT     size=2>about 40ms.  I don't know why, they are scsi drives and there     isn't that much stress on the syste.  Anyway, I     moved the redo logs to seperate disks, but I fear I     may have made an error.  Instead of moving each "group" to     seperate hard disks, I moved the members of each group to     seperate hard disks.  Anyway, I am wondering if     anyone has advice on this problem.
    Sincerely, Kevin Kostyszyn
    DBA Dulcian, Inc <FONT
    size=2>www.dulcian.com kevin_at_dulcian.com

Received on Thu Mar 22 2001 - 11:30:37 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US