Exploring Onstat

- an IBM Informix IDS utility

Lester Knutsen
Advanced DataTools Corporation
Onstat – monitor IDS server
operations
• Onstat utility reads shared-memory structures
and provides statistics about the database
server at the time that the command executes.
• The contents of shared memory might change
as the onstat output displays.
• The onstat utility does not place any locks on
shared memory, so running the utility does not
affect performance.
• Onstat is a key utility to monitor the
performance of your IDS server

2
 Informix IDS Shared Memory

•Control Tables
Resident Memory Segment •Buffers
LRU Page Page

LRU Page Page

LRU Page Page

LRU Page Page

•Dictionary cache
Virtual Memory Segment •Working Storage
•Sort Space

Message Memory Segment

3
Onstat options:
onstat [ -abcdfghklmpstuxzBCDFRX ]
[ -i ]
[ -r [<seconds>] ]
[ -o [<outfile>] ] [ <infile> ]
-i Interactive mode
-r Repeat every <seconds> seconds (default: 5)
-o Put shared memory into specified dump file
<infile> Read shared memory information from specified
dump file
-z Zero profile counts

4
Key Onstat options:
• -d Print spaces and chunks
• -k Print locks
• -l Print logging
• -m Print message log
• -p Print profile
• -T Print TBLspaces
• -u Print user threads
• -D Print spaces and detailed chunk stats
• -F Print page flushers
• -R Print LRU queues
5
Key Onstat –g options
(New monitoring commands)
• ath Print all threads
• rea Print ready threads
• sch Print VP scheduler statistics
• seg Print memory segment statistics
• iof Print disk IO statistics by chunk/file
• ses [<session id>]Print session information
• sql [<session id>]Print SQL information
• mgm Print Memory Grant Manager information

6
Current status of Server: onstat -

Current status: onstat -

lester@merlin >onstat -

INFORMIX-OnLine Version 9.2X.XXX-- On-Line -- Up 7 days

11:54:44 -- 10656 Kbytes

Current status when Server is down

lester@merlin >onstat -
shared memory not initialized for INFORMIXSERVER 'merlindb'
lester@merlin >

7
Onstat Header Information
Informix Dynamic Server 2000 Version 9.21.UC6 -- On-Line -- Up 2 days 23:48:57 --
3694592 Kbytes

• Product and Version

• Mode (and Type)
• (Optional Reason when Server is Blocked)
• Time Server has been up
• Size of Shared Memonry in Kbytes

8
Mode of Server
• Off-Line Mode (does not show in header)
• Quiescent Mode
• On-Line Mode
• Read-Only Mode (DR Only)
• Recovery Mode
• Shutdown Mode

9
Reason when Server is blocked
• CKPT - Checkpoint
• LONGTX - Long transaction
• ARCHIVE - Ongoing storage-space backup
• MEDIA_FAILURE - Media failure
• HANG_SYSTEM - Database server failure
• DBS_DROP - Dropping a dbspace
• DDR - Discrete data replication (IDS)
• LBU - Logs full high-watermark

10
Server Profile:
onstat -p
Informix Dynamic Server 2000 Version 9.21.UC6 -- On-Line -- Up 2 days 23:48:57 --
3694592 Kbytes
Profile
dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached
412782302 493730321 2917750938 85.85 80916102 88592926 705858905 88.54

isamtot open start read write rewrite delete commit rollbk

1282651515 31183189 128955107 4056268803 334566515 47983470 4742621 125242 6

gp_read gp_write gp_rewrt gp_del gp_alloc gp_free gp_curs

3 0 0 0 0 0 2

ovlock ovuserthread ovbuff usercpu syscpu numckpts flushes

0 0 0 286397.69 18929.10 1382 3210

bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress seqscans

14692696 10 1065062633 0 0 2151 10377337 4240282

ixda-RA idx-RA da-RA RA-pgsused lchwaits

12294388 8635422 309229639 330153243 31011288

11
 Informix IDS Shared Memory
•Control Tables
Resident Memory Segment •Buffers
LRU Page Page

LRU Page Page

LRU Page Page

LRU Page Page

•Dictionary cache
Virtual Memory Segment •Working Storage
•Sort Space

Message Memory Segment

Page

12
 Page Gets Read into Memory by a Select
•Control Tables
Resident Memory Segment •Buffers
LRU Page Page

LRU Page Page

LRU Page Page

LRU Page Page

•Dictionary cache
Virtual Memory Segment •Working Storage
•Sort Space

Message Memory Segment

Page

13
When all Buffers are full, Least Recently Used (LRU)
Page is discarded to make room for more Data
•Control Tables
Resident Memory Segment •Buffers
LRU Page Page

LRU Page Page

LRU Page Page

LRU Page Page

•Dictionary cache
Virtual Memory Segment •Working Storage
•Sort Space

Message Memory Segment

Page Page
Page Page Page
Page
14
 When a user updates a page, it is marked as dirty and must be
written out to disk before it is discarded.
•Control Tables
Resident Memory Segment •Buffers
Updates
LRU Page Page
UpdatesLRU Page Page

UpdatesLRU Page Page

LRU Page Page
Updates

•Dictionary cache
Virtual Memory Segment •Working Storage
•Sort Space

Message Memory Segment

Page Page
Page Page Page
Page
15
 LRU Queues manage writing data to disk in the background
when there is idle time based on LRU_MAX_DIRTY and
LRU_MIN_DIRTY ONCONFIG
Resident Memory values •Control Tables
Segment •Buffers
Updates
LRU Page Page
UpdatesLRU Page Page

UpdatesLRU Page Page

LRU Page Page
Updates

•Dictionary cache
Virtual Memory Segment •Working Storage
•Sort Space

Message Memory Segment

Page Page
Page Page Page
Page
16
 When all buffers are Dirty, the server must STOP all processing and
perform a Foreground Write
•Control Tables
Resident Memory Segment •Buffers
Updates
LRU Page Page
All Activity is Blocked

UpdatesLRU Page Page

UpdatesLRU Page Page

LRU Page Page
Updates

•Dictionary cache
Virtual Memory Segment •Working Storage
•Sort Space

Message Memory Segment

Page Page
Page Page Page
Page
17
 When a buffer is written to disk, it is marked as clean and may
be discarded if needed.
•Control Tables
Resident Memory Segment •Buffers
Updates
LRU Page Page
UpdatesLRU Page Page

UpdatesLRU Page Page

LRU Page Page
Updates

•Dictionary cache
Virtual Memory Segment •Working Storage
•Sort Space

Message Memory Segment

Page Page
Page Page Page
Page
18
 A Standard (7.X) Checkpoint writes all Dirty Buffers to Disk

•Control Tables
Resident Memory Segment •Buffers
Updates
LRU Page Page
UpdatesLRU Page Page

UpdatesLRU Page Page

LRU Page Page
Updates

•Dictionary cache
Virtual Memory Segment •Working Storage
•Sort Space

Message Memory Segment

Page Page
Page Page Page
Page
19
 A Fuzzy (9.21) Checkpoint flags all Dirty Buffers and writes them to
Disk in the background without blocking activity
•Control Tables
Resident Memory Segment •Buffers
Updates
LRU Page Page
UpdatesLRU Page Page

UpdatesLRU Page Page

LRU Page Page
Updates

•Dictionary cache
Virtual Memory Segment •Working Storage
•Sort Space

Message Memory Segment

Page Page
Page Page Page
Page
20
After a Checkpoint, all Buffers are Clean and the cycle
Starts over again
•Control Tables
Resident Memory Segment •Buffers
LRU Page Page

LRU Page Page

LRU Page Page

LRU Page Page

•Dictionary cache
Virtual Memory Segment •Working Storage
•Sort Space

Message Memory Segment

Page Page
Page Page Page
Page
21
Key Elements of onstat -p

• Reads %cached - The goal is t> 95%

• Writes %cached - The goal is >85%
• The BUFFERS parameter in your ONCONFIG file will effect this value.
• Be careful, if you make the BUFFERS too large this will take memory away from
other processes and may slow down your whole system.
• ovlock - This should be zero. Configure using LOCKS parameter in the ONCONFIG file.
• ovuserthread - This should be zero. Configure number of user threads in the third value
of the NETTYPE parameter
• ovbuff - This should be zero. Configure using the BUFFER parameter in the
ONCONFIG file.
• bufwaits - This should be zero. This indicates the number of times a user thread has
waited for a BUFFER.
• lokwaits - This should be zero. This indicates the number of times a user thread has
waited for a LOCK.
• deadlks - This should be zero. This indicates the number of times a deadlock was
detected and prevented.
• dltouts - This should be zero. This indicates the number of times a distributed deadlock
was detected.

22
Key Ratios Calculated from
Onstat -p
• Kbytes read and written per minute/hour
• Buffer turnover ratio per minute/hour
• Buffer wait ratio
• Read Ahead Utilization

23
Display message log file:
onstat -m
lester@merlin >onstat -m
INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 7 days 12:41:12 -- 10656
Kbytes
Message Log File: /u3/informix/online1.log
21:46:51 Checkpoint Completed: duration was 8 seconds.
21:52:00 Checkpoint Completed: duration was 9 seconds.
21:57:09 Checkpoint Completed: duration was 8 seconds.
22:00:42 Logical Log 20 Complete.
22:00:43 Process exited with return code 1: /bin/sh /bin/sh -c
/u3/informix/log_full.sh 2 23 "Logical Log 20 Complete." "Logical Log 20 Complete."
22:02:17 Checkpoint Completed: duration was 8 seconds.

24
Using “tail -f” to continuously
show the end of message log file
• Note: I like to have the OnLine log file always display in one of
my windows on screen. The trick to doing this is to use the
UNIX "tail" command with the "-f" option. This continually reads
the last lines of a file as it is appended to. On my system I run
the following command to continually monitor this log:

tail -f $INFORMIXDIR/online.log

25
User status: onstat -u

Informix Dynamic Server 2000 Version 9.21.UC6 -- On-Line -- Up 2 days 23:50:

12 -- 3694592 Kbytes
Userthreads
address flags sessid user tty wait tout locks nreads nwrites
5911e018 ---P--D 1 informix - 0 0 0 6581 270848
5911e60c ---P--F 0 informix - 0 0 0 0 3412274
5911ec00 ---P--F 0 informix - 0 0 0 0 2407535
59e4460c ---P--- 12 informix - 0 0 0 0 20434
59e44c00 ---P--B 13 informix - 0 0 0 11 258
59e457e8 --AP--M 3180 root - 0 0 0 114 109
59e45ddc ---P--D 16 informix - 0 0 0 0 0
59e463d0 Y-AP--M 3191 root - 69542450 0 0 113 109
59e469c4 Y------ 2621 adtdba - 6eb263d0 0 2 0 0
59e4ecbc ------- 3184 root - 0 0 0 0 0
59e4f8a4 Y------ 2621 adtdba - 5bf46b20 0 2 2742102 0
59e50a80 Y------ 2621 adtdba - 5c1ee948 0 3 0 0
59e53a20 --AP--M 3186 root - 0 0 0 116 109
59e551f0 ------- 3186 root - 0 0 0 0 0
59e57b9c Y------ 2621 adtdba - 5c973850 0 2 0 0
59e5cef4 Y------ 2621 adtdba - 5c2c4af8 0 3 0 0
59e5d4e8 ---P--M 3192 root - 0 0 1 0 88

26
User status: onstat -u Flags
Flags in position 1
B - Waiting on a buffer
C - Waiting on a checkpoint
G - Waiting on a logical log buffer write
L - Waiting on a lock
S - Waiting on a mutex
T - Waiting on a transaction
Y - Waiting on a condition
X - Waiting on a transaction rollback
Flags in position 2
* - Transaction active during I/O error
Flags in position 3
A - Dbspace backup thread
B - Begin work
P - Prepared for commit work
X - TP/XA prepared for commit work
C - Committing work
R - Rolling back work
H - Heuristically rolling back work

27
User status: onstat -u Flags

Flags in position 4
P - Primary thread for a session
Flags in position 5
R - Reading call
X - Transaction is committing
Flags in position 6
None
Flags in position 7
B - Btree cleaner thread
C - Cleanup of terminated user
D - Daemon thread
F - Page flusher thread
M - ON-Monitor user thread

28
Logical Logs status: onstat -l

Informix Dynamic Server 2000 Version 9.21.UC6 -- On-Line -- Up 2 days 23:52:

51 -- 3694592 Kbytes

Physical Logging
Buffer bufused bufsize numpages numwrits pages/io
P-1 8 128 34175115 268502 127.28
phybegin physize phypos phyused %used
10003f 500000 108581 8 0.00

Logical Logging
Buffer bufused bufsize numrecs numpages numwrits recs/pages pages/io
L-3 0 64 1419836 62228 8468 22.8 7.3
Subsystem numrecs Log Space used
OLDRSAM 1419836 113663544

address number flags uniqid begin size used %used

a8d8d70 25 U-B---- 3769 10400035 5000 5000 100.00
a8d8d8c 26 U-B---- 3770 104013bd 5000 5000 100.00
a8d8da8 27 U-B---- 3771 10402745 5000 5000 100.00
a8d8dc4 28 U-B---- 3772 10403acd 5000 5000 100.00
a8d8de0 29 U-B---- 3773 10404e55 5000 2397 47.94

29
Logical Logs status:
onstat -l Flags
The flags column provides status information about each log.
A - Newly added, must run an archive before they can be used
B - Backed up to tape or "/dev/null"
C - Current logical log file
F - Free and available for use. You will rarely see this flag
as logs are not marked as free until right before they
are needed.
L - Last checkpoint is in this logical log
U - Used logical log, it may be free if it is backed up and
contains no active transactions.

One way to tell which logs can be reused is to use "onstat -l" with
"onstat -x" to display all active sessions.

30
Display transactions: onstat -x

This option displays all current transactions. The most useful column is "log
begin". This tells you in which logical log a transaction started. This
may be used with the "onstat -l" command to determine which logs are free
and may be reused. Find the earliest logical log number in the column "log
begin". This tells you which logical log has the earliest active
transaction. Any logical logs that are backed up before the log with the
earliest transaction will be automatically reused by OnLine.

lester@merlin >onstat -x
INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 01:21:38 -- 10656
Transactions
address flags userthread locks log begin isolation retrys coordinator
a2f4384 A---- a2d1118 0 0 NOTRANS 0
a2f44a8 A---- a2d1558 0 0 COMMIT 0
a2f45cc A-B-- a2d1118 2 21 NOTRANS 0
6 active, 128 total, 7 maximum concurrent

31
Display locks: onstat -k

WARNING: If you have a large number of LOCKS defined in your ONCONFIG file and many
users you could see thousands of rows from this command.

lester@merlin >onstat -k
INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 7 days 12:53:31 -- 10656
Locks
address wtlist owner lklist type tblsnum rowid key#/bsiz
a103e74 0 a2d2218 0 HDR+S 100002 20a 0
1 active, 20000 total, 16384 hash buckets

32
Who owns a lock

• The "owner" column lists the address in shared memory of

the user who owns a lock. Use this with "onstat -u" to see
all users, and compare this with the "address" column to
identify username of the owner.

33
What table is locked?

• The "tblsnum" column identifies the table that is being locked.

Compare this with the output of the following SQL statement to
convert a table's partnum to hex. This will identify which table
is locked.

1. Find a list of tblsnum

dbaccess database - <<EOF
select tabname, hex(partnum) tblsnum
from systables where tabid > 99;
EOF

database selected

tabname tblsnum
genjournal 0x0010009E
gjsum 0x0010009F

34
What table is locked?

2. Find what is locked

onstat -k

INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 01:47:38 -- 10656

Locks
address wtlist owner lklist type tblsnum rowid key#/bsiz
a103e44 0 a2d1118 a103de4 HDR+X 10009f 0 0
3 active, 20000 total, 16384 hash buckets

3. Compare tblsnum from step 1 and step 2.

This identifies the table gjsum as the one that is locked.

• The tblsnum 100002 has a special meaning. This indicates

a database lock. Every user who opens a database will
place a shared lock on the database.

35
Types of locks

• Database - Lock is placed on tablespace 1000002

• Table - Lock is placed on actual tablespace with rowid of 0
• Page - Lock is placed on tablespace with rowid ending in 00
• Row - Lock is placed on tablespace with actual rowid (not 00)
• Byte - Lock is placed on tablespace/page with size of bytes
• Key - Lock is placed on tablespace hex rowid (starting with f)

36
Types of locks Flags

HDR - Header
B - Bytes lock
S - Shared lock
X - Exclusive
I - Intent
U - Update
IX - Intent-exclusive
IS - Intent-shared
SIX - Shared, Intent-exclusive

37
Dbspaces and chunks status:
onstat -d
lester@merlin >onstat -d

INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 7 days 12:54:44 -- 10656

Kbytes

Dbspaces
address number flags fchunk nchunks flags owner name
a2ce100 1 1 1 1 N informix rootdbs
a2ce508 2 1 2 1 N informix dbspace1
a2ce578 3 1 3 1 N informix dbspace2
a2ce5e8 4 1 4 1 N informix dbspace3
4 active, 2047 maximum

Chunks
address chk/dbs offset size free bpages flags pathname
a2ce170 1 1 0 250000 62047 PO- /u3/dev/rootdbs1
a2ce280 2 2 0 10000 9587 PO- /u3/dev/dbspace1
a2ce358 3 3 0 10000 9947 PO- /u3/dev/dbspace2
a2ce430 4 4 0 10000 9947 PO- /u3/dev/dbspace3
4 active, 2047 maximum

38
onstat -d Flags

The "flags" for Dbspaces are: The "flags" for Chunks are:
Position 1 Position 1
M - Mirrored Dbspace P - Primary
N - Not Mirrored Dbspace M - Mirror
Position 2 Position 2
X - Newly mirrored O - On-line
P - Physical recovery D - Down
underway
X - Newly mirrored
L - Logical recovery underway
I - Inconsistent
R - Recovery underway
Position 3
Position 3
B - Blobspace Dbspace
B - Blobspace
T - Temporary Dbspace

39
Disk I/O: onstat -D

lester@merlin >onstat -D

INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 7 days 12:55:09 -- 10656

Kbytes

Dbspaces
address number flags fchunk nchunks flags owner name
a2ce100 1 1 1 1 N informix rootdbs
a2ce508 2 1 2 1 N informix dbspace1
a2ce578 3 1 3 1 N informix dbspace2
a2ce5e8 4 1 4 1 N informix dbspace3
4 active, 2047 maximum

Chunks
address chk/dbs offset page Rd page Wr pathname
a2ce170 1 1 0 36563 179558 /u3/dev/rootdbs1
a2ce280 2 2 0 3 0 /u3/dev/dbspace1
a2ce358 3 3 0 2 0 /u3/dev/dbspace2
a2ce430 4 4 0 2 0 /u3/dev/dbspace3
4 active, 2047 maximum

40
Writing to Disk - onstat -F

lester@merlin >onstat -F

INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 7 days

Fg Writes LRU Writes Chunk Writes

168 172280 5277

address flusher state data

a2d0458 0 I 0 = 0X0
states: Exit Idle Chunk Lru

41
Writing to Disk - onstat -F

• Foreground writes occur when the Server needs a buffer and must
interrupt processing to flush buffers to disk to free a buffer. These
are the least desirable type of writes.

• Background writes (LRU Writes) occur when a set percent of the

buffers are dirty. This is controlled by the LRU parameters in the
ONCONFIG file. These do not interrupt user processing and are
the best for interactive systems.

• Chunk writes occur at checkpoints, and all dirty buffer pages are
written to disk. The more dirty pages, the longer a checkpoint will
take. Checkpoint writes are sorted and optimized, but the longer a
checkpoint is, the longer it will block user activity. Checkpoint
writes are best for batch systems.

42
List all threads: onstat -g ath

INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 7 days 12:56:09 -- 10656

Threads:
tid tcb rstcb prty status vp-class name

2 a336b70 0 2 sleeping(Forever) 3lio lio vp 0

3 a336dd0 0 2 sleeping(Forever) 4pio pio vp 0
4 a337088 0 2 sleeping(Forever) 5aio aio vp 0
5 a337340 0 2 sleeping(Forever) 6msc msc vp 0
6 a337af8 0 2 sleeping(Forever) 7aio aio vp 1
7 a337e00 a2d0018 4 sleeping(secs: 1) 1cpu main_loop()
8 a34ab48 0 2 running 1cpu sm_poll
9 a34b770 0 2 running 8tli tlitcppoll
10 a34bce0 0 2 sleeping(Forever) 1cpu sm_listen
11 a3c4a28 0 2 sleeping(secs: 2) 1cpu sm_discon
12 a3c4e58 0 3 sleeping(Forever) 1cpu tlitcplst
13 a3d0680 a2d0458 2 sleeping(Forever) 1cpu flush_sub(0)
14 a3d0e40 a2d0898 2 sleeping(secs: 8) 1cpu btclean
30 a35ea58 a2d1558 4 sleeping(secs: 1) 1cpu onmode_mon
283 a39ef38 a2d2218 2 cond wait(sm_read) 1cpu sqlexec

43
List Virtual Processor status:
onstat -g sch
This option provides the means to identify which "oninit" UNIX process
corresponds to which Informix Server server Virtual Processor.

lester@merlin >onstat -g sch

INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 7 days 12:56:46 -- 10656

Kbytes

VP Scheduler Statistics:

vp pid class semops busy waits spins/wait

1 230 cpu 21 0 0
2 231 adm 0 0 0
3 232 lio 277 0 0
4 233 pio 62 0 0
5 234 aio 144794 0 0
6 235 msc 756 0 0
7 236 aio 64028 0 0
8 237 tli 3 0 0

44
List SQL statement types:
onstat -g sql
• This is the most interesting of the new options. This option allows you to
drill down and see the actual SQL statement that a user is executing.

lester@merlin >onstat -g sql

INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 7 days 12:52:02 --

Sess SQL Current Iso Lock SQL ISAM F.E.
Id Stmt type Database Lvl Mode ERR ERR Vers
264 INSERT ffsdw NL Not Wait -264 0 9.10

45
List SQL statement for a user:
onstat -g sql sid
lester@merlin >onstat -g sql 264

INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 7 days 12:51:10 -- 10656

Kbytes

Sess SQL Current Iso Lock SQL ISAM F.E.

Id Stmt type Database Lvl Mode ERR ERR Vers
264 INSERT ffsdw NL Not Wait -264 0 9.10

Current SQL statement :

insert into gjsum select exp_org, exp_prog, bud_obj_code, job_num,
sum (exp_amount) from genjournal group by 1, 2, 3, 4

Last parsed SQL statement :

insert into gjsum select exp_org, exp_prog, bud_obj_code, job_num,
sum (exp_amount) from genjournal group by 1, 2, 3, 4

46
List users sessions: onstat -g ses

• This option shows additional information about users' sessions, including

how much memory each session is using.

lester@merlin >onstat -g ses

INFORMIX-OnLine Version 9.2X.XXX -- On-Line -- Up 7 days 12:57:48 -- 10656

Kbytes

session #RSAM total used

id user tty pid hostname threads memory memory
265 informix - 0 - 0 8192 4680
264 lester 4 4249 merlin 1 106496 97840
10 informix - 0 - 0 8192 4680
7 informix - 0 - 0 16384 13144
6 informix - 0 - 0 8192 4680
4 informix - 0 - 0 16384 13144
3 informix - 0 - 0 8192 4680
2 informix - 0 - 0 8192 4680

47
Other Onstat Options

Repeat ONSTAT commands: -r

• To continually repeat an ONSTAT command use the "-r # of
seconds" option. This is very useful when you need to monitor a
situation. The following example displays the status of the logical
logs every 10 seconds.

onstat -l -r 10

Clear ONSTAT shared memory statistics: onstat -z

• The Server statistics are reset every time OnLine is restarted. To
reset all the statistics while OnLine is running, without shutting it
down, use the following command:

onstat -z

48
For more information
• International Informix User Group Software
Repository – http://www.iiug.org
• Mark Scranton’s web site
http://www.markscranton.com/informix
• IBM Informix Documentation
• IDS Administrators Reference Manual

49
Advanced Training Options
• Advanced Informix IDS Performance Tuning
and Optimization
– March 31 - April 4, 2003
– An advanced course for experienced Informix
DBAs.
• Informix IDS for Database Administrators
– February 10-14, 2003
– Configuring, managing, and tuning the Informix IDS
database server.

50
 Lester Knutsen
Advanced DataTools Corporation
Web site: www.advancedatatools.com

(download SQL scripts and papers)

Email: [email protected]

Phone: 703-256-0267 x102

Contact information:

Lester Knutsen
Advanced DataTools Corporation

Web site: www.advancedatatools.com

(download SQL scripts and paper)

Email: [email protected]

Phone: 703-256-0267 x102

52