RE: performance with very active threads
Tam
Comments below on onconfig.
Is KAIO available on SCO Unixware? Your AIO maxlen look long to me.
The problem looks to me in the application. The engine must wait for all
threads to get out of critical sections before starting the checkpoint, but
then there is the time taken to clean all buffers etc on the checkpoint.
The former is often the longer in duration.
Is you disk I/O maxing out, particularly at checkpoints, and on which
drives?
Is is I/O even between your disk drives? If not, move tables to even the
load.
MW
-----Original Message-----
From: Tam McLaughlin
Sent: Thursday, September 07, 2000 2:05 AM
To: informix-list@iiug.org
Subject: Re: performance with very active threads
I forgot to include my onconfig file. From a posting i read the other day, I
can see there a few things
I can do the next time I restart the engine: unset TBLSPACE_STATS and reduce
my PHYSFILE size.
PHYSDBS rootdbs Move elsewhere? Logical logs also!
PHYSFILE 10000 Increase, for increased checkpoint interval
LOGFILES 21
LOGSIZE 5000TBLSPACE_STATS 1 Set off unless you really need them
TAPEDEV /dev/rmt/ctape3
TAPEBLK 32
TAPESIZE 8000000LTAPEDEV /dev/rmt/ctape2
LTAPEBLK 16
LTAPESIZE 525000STAGEBLOB
SERVERNUM 0
DBSERVERNAME pluto
DBSERVERALIASES pluto_net
NETTYPE ipcshm,3,64,CPU
NETTYPE tlitcp,1,10,NET
DEADLOCK_TIMEOUT 60
RESIDENT 0
MULTIPROCESSOR 1
NUMCPUVPS 3
SINGLE_CPU_VP 0
NOAGE 0
AFF_SPROC 0
AFF_NPROCS 0
LOCKS 20000 BUFFERS 96000 Far too few. How much memory do you have?
NUMAIOVPS 8 May want to increase - double
PHYSBUFF 120
LOGBUFF 26LOGSMAX 60
CLEANERS 8
SHMBASE 0xa000000SHMVIRTSIZE 120000 Check is this used? If not move memory to buffers.
SHMADD 32000
SHMTOTAL 0CKPTINTVL 300 Suggest you increase - 30 mins suggested
LRUS 63
LRU_MAX_DIRTY 2
LRU_MIN_DIRTY 1
LTXHWM 40
LTXEHWM 45
TXTIMEOUT 0x12c
STACKSIZE 32
OFF_RECVRY_THREADS 10
ON_RECVRY_THREADS 1
DRAUTO 0
DRINTERVAL 30
DRTIMEOUT 30
CDR_LOGBUFFERS 2048
CDR_EVALTHREADS 1,2
CDR_DSLOCKWAIT 5
CDR_QUEUEMEM 4096
BAR_ACT_LOG /tmp/bar_act.log
BAR_MAX_BACKUP 0
BAR_RETRY 1
BAR_NB_XPORT_COUNT 10
BAR_XFER_BUF_SIZE 31ISM_DATA_POOL ISMData
ISM_LOG_POOL ISMLogs
RA_PAGES Set both of these
RA_THRESHOLD You may set to 0 with too few buffers
DBSPACETEMP tempspace Try 3 on different disks
DUMPDIR /ascii
DUMPSHMEM 1
DUMPGCORE 0
DUMPCORE 0
DUMPCNT 1
FILLFACTOR 90
USEOSTIME 0
MAX_PDQPRIORITY 100
DS_MAX_QUERIES
DS_TOTAL_MEMORY
DS_MAX_SCANS 1048576
DATASKIP off
OPTCOMPIND 0
ONDBSPACEDOWN 2
LBU_PRESERVE 1OPCACHEMAX 0
HETERO_COMMIT 0
OPT_GOAL -1
DIRECTIVES 1
RESTARTABLE_RESTORE off
Murray Wood wrote in message <8p6he5$ct5$1@news.xmission.com>...
>
>Tam
>
>Comments below on onconfig.
>
>Is KAIO available on SCO Unixware? Your AIO maxlen look long to me.
KAIO is not supported on unixware 7.0.1. We will be upgrading the OS in a
few
months and will also upgrade to IDS2000
>
>The problem looks to me in the application. The engine must wait for all
>threads to get out of critical sections before starting the checkpoint, but
>then there is the time taken to clean all buffers etc on the checkpoint.
>The former is often the longer in duration.
>
>Is you disk I/O maxing out, particularly at checkpoints, and on which
>drives?
>Is is I/O even between your disk drives? If not, move tables to even the
>load.
>
Not sure what you mean but using sar, the %wio is > 50.
Moving from RAID5-RAI10 will, I believe, help.
It was mentioned to increase the physfile size, but at checkpoints, it is
always < 5%.
>PHYSDBS rootdbs Move elsewhere? Logical logs also!
I intend to do both these
>PHYSFILE 10000 Increase, for increased checkpoint interval
> BUFFERS 96000 Far too few. How much memory do you have?
756 Mb RAM
>>SHMVIRTSIZE 120000 Check is this used? If not move memory to buffers.
i have this:
Informix Dynamic Server Version 7.30.UC10X2 -- On-Line -- Up 9 days
06:13:34 -- 323080 Kbytes
Segment Summary:
id key addr size ovhd class blkused blkfree
1003 1381386241 a000000 207953920 4032 R 25378 7
1004 1381386242 16652000 122880000 2472 V 2496 12504
1005 1381386243 1db82000 753664 608 M 87 5
1006 1381386244 1dc3a000 753664 608 M 84 8
1007 1381386245 1dcf2000 753664 608 M 84 8
Total: - - 333094912 - - 28129 12532
>CKPTINTVL 300 Suggest you increase - 30 mins suggested
>LRUS 63> >>RA_PAGES Set both of these
>RA_THRESHOLD You may set to 0 with too few buffers
>DBSPACETEMP tempspace Try 3 on different disks
>>>