IBM WebSphere MQ for z/OS V8 - Latest Features Deep Dive

Z1: WebSphere MQ for z/OS V8:
Latest Features Deep Dive
Damon Cross, Advisory Software Engineer
damon_cross@uk.ibm.com

© © 2014 IBM C 2or0p1o4ra ItBioMn Corporation
Please Note
IBM’s statements regarding its plans, directions, and intent are subject to change or withdrawal
without notice at IBM’s sole discretion. Information regarding potential future products is
intended to outline our general product direction and it should not be relied on in making a
purchasing decision.
The information mentioned regarding potential future products is not a commitment, promise,
or legal obligation to deliver any material, code or functionality. Information about potential
future products may not be incorporated into any contract. The development, release, and
timing of any future features or functionality described for our products remains at our sole
discretion
Performance is based on measurements and projections using standard IBM benchmarks in a
controlled environment. The actual throughput or performance that any user will experience
will vary depending upon many factors, including considerations such as the amount of
multiprogramming in the user’s job stream, the I/O configuration, the storage configuration, and
the workload processed. Therefore, no assurance can be given that an individual user will
achieve results similar to those stated here.

© 2014 IBM Corporation
3
Agenda
 64 Bit Buffer Pools
 8 Byte Log RBA
 Chinit SMF
 SCM Storage (Flash)
 Other things…

Z1: IBM WebSphere MQ for z/OS
64-bit Buffer Pools

5
Buffer pools – Background
Q1
4KB Page
Page Set
Q2
Page Set

7
Buffer pools – Using only 31-bit storage
DATA
DATA
CODE
16 EB
2 GB BAR
Queue Manager Address Space
Buffer Pool Buffer Pool Buffer Pool Buffer Pool Max 1.6GB for
Buffer Pools

9
Buffer pools – The problems
 Not much space below the bar for buffer pools
- Maximum 1.6GB, depending on common area
 Put/Get with bufferpool = 'memory' speed
 Put/Get from page set = 'disk' speed
 Can spend a lot of time moving data around
- Getting pages from page set into buffer pool
- Putting pages from buffer pool into page set
- This is detrimental for performance
 A maximum of 16 buffer pools
- Although up to 100 page sets are supported
 Lots of time spent performing tuning
- buffer pool sizes and queue/buffer pool/page set mappings

11
64 Bit Buffer Pools: The Solution
 Allow buffer pools to be above the bar.
- Buffer pools can (theoretically) make use of up to 16 EB
of storage
 Increase maximum size of buffer pool
- if above the bar
 Allow up to 100 buffer pools
- Can have a 1-1 page set to buffer pool mapping

13
Buffer pools – Using 64-bit storage
DATA
DATA
CODE
2 GB BAR
Queue Manager Address Space
Buffer Pool
Buffer Pool
16 EB
Buffer Pool Buffer Pool Buffer Pool

15
64 Bit Buffer Pools: What has changed?
DEFINE BUFFPOOL(<id>)
BUFFERS(<integer>
PAGECLASS(4KB/FIXED4KB)
LOCATION(BELOW/ABOVE) NOREPLACE/REPLACE
- BUFFPOOL id
- 0 to 99
- BUFFERS integer
- Up to 500,000 if LOCATION(BELOW)
- Up to 999,999,999 if LOCATION(ABOVE)
- PAGECLASS can be:
- 4KB, if LOCATION(BELOW)
- FIXED4KB, if LOCATION(ABOVE)
- permanent backing by real storage for life of Queue Manager
- no need to programmatically page fix/unfix when doing I/O
- better performance
- ensure you have enough real storage available
- LOCATION
- BELOW – buffer pool is below the bar (default)
- ABOVE – buffer pool is above the bar

17
Buffer pools – Migration
•To use buffer pools above the bar version 8 new
function must be enabled using
OPMODE(NEWFUNC,800)
–If the queue manager is running in compatibility mode
the behaviour is the same as in previous releases
•Specifying LOCATION(BELOW) is valid regardless
of OPMODE
•Some console messages have changed regardless of
the value of OPMODE

19
Buffer pools – Configuration
• CSQINP1
– DEFINE BUFFPOOL(22) LOCATION(ABOVE)
BUFFERS(1024000) REPLACE
– DEFINE BUFFPOOL(88) BUFFERS(12000) REPLACE
• CSQINP1 or dynamically with DSN
– DEFINE PSID(22) BUFFPOOL(22) REPLACE
• CSQINP2 or dynamically
– ALTER BUFFPOOL(88) LOC(ABOVE)
CSQP024I !MQ21 Request initiated for buffer pool 88
CSQ9022I !MQ21 CSQPALTB ' ALTER BUFFPOOL' NORMAL COMPLETION
CSQP023I !MQ21 Request completed for buffer pool 88, now has 12000 buffers
CSQP054I !MQ21 Buffer pool 88 is now located above the bar

20
Buffer pools – Configuration
• DISPLAY USAGE PSID(*)
CSQI010I !MQ21 Page set usage …
<REMOVED>
End of page set report
CSQI065I !MQ21 Buffer pool attributes ...
Buffer Available Stealable Stealable Page Location
pool buffers buffers percentage class
_ 0 1024 1000 99 4KB BELOW
_ 22 1024000 234561 23 FIXED4KB ABOVE
_ 88 12000 1200 10 4KB ABOVE
End of buffer pool attributes

21
Buffer pools – Performance
• 16 CP LPAR
• Each transaction puts and gets a random message from a pre-loaded queue. The
second test requires a doubling in buffer pool size.

8-Byte Log RBA

25
Log RBA – The problem
• WebSphere MQ for z/OS V7.1 (or prior):
– Uses a 6-byte log RBA (Relative Byte Address)
– RBA range of 0 to x'FFFFFFFFFFFF' (= 255TB)
– Some customers reach this limit in 12 – 18 months
– At 100MB/sec the log would become full in 1 month
• If the end of the log RBA range is reached:
– Queue manager terminates and requires a cold start
• Disruptive outage!
– Potential for loss of persistent data
• To avoid an outage:
– Run CSQUTIL RESETPAGE at regular planned intervals to reset the log
RBA, and define new BSDS and active log datasets.

26
Warning Messages and abend
 V7.1 Queue Managers do issue warning messages as log RBA gets high:
CSQI045I when log RBA is x'700000000000', x'7100..', x'7200..' and
x'7300..'
CSQI046E when log RBA is x'740000000000', x'7500..', x'7600..' and
x'7700..'
CSQI047E when log RBA is x'780000000000', x'7900..', x'nn00..' and
x'FF00..'
 APAR PM48299 (WebSphere MQ V7.0.1 and above) added messages:
 CSQJ032E when log RBA is higher than x'F80000000000'
CSQJ031D to confirm restart even though log RBA has passed
x'FF8000000000'
 To prevent loss of data, Queue Managers with APAR PM48299 applied:
 Terminate with abend 00D10257 when log RBA reaches x'FFF800000000'

29
Log RBA – The solution
• IBM WebSphere MQ for z/OS V8.0 supports a 8-byte log RBA
– Upper limit on logical log is 64K times larger
– At 100MB/sec the log would become full in 5578 years!
• The BSDS and log records have been updated to support the larger 8-
byte RBAs and URIDs
• Utilities or applications that read the following are impacted:
– BSDS
– Logs
– Console messages that contain the log RBA or URID
• Queue managers use a 6-byte RBA until 8-byte RBA is enabled
– BSDS conversion utility to migrate to 8-byte RBA (same model that is
used by DB2)

31
Log RBA – Enabling 8-byte log RBA
• Version 8 new function must be enabled
– In a QSG new function is only enabled once ALL queue managers
have been started with OPMODE(NEWFUNC,800)
• Stop the queue manager
• Run the new BSDS conversion utility (CSQJUCNV)
– Converts version 1 input BSDS data set(s) to version 2
–Writes converted BSDS to new data set(s)
• Rename the BSDS data sets, or change the queue manager
started task JCL to use the new BSDS data sets
• Restart the queue manager

BSDS conversion utility (CSQJUCNV)
 Parameters
- NOQSG (specify for a stand alone queue manager)
 No OPMODE checks performed
- INQSG,qsgname,dsgname,db2ssid (specify for a queue manager in a QSG)
 Utility checks that all QMgrs in the QSG have been started at
OPMODE(NEWFUNC,800) before allowing conversion to proceed
 Example JCL:
//CSQ4BCNV JOB
//CONVERT EXEC
PGM=CSQJUCNV,REGION=32M,PARM=('INQSG,SQ13,DB2,DB4A')
//STEPLIB DD DSN=ANTZ.MQ.V000.CUR.SCSQAUTH,DISP=SHR
// DD DSN=ANTZ.MQ.V000.CUR.SCSQANLE,DISP=SHR
// DD DSN=SYS2.DB2.V10.SDSNLOAD,DISP=SHR
//SYSPRINT DD SYSOUT=*
//SYSUT1 DD DSN=VICY.MQ1O.BSDS01,DISP=SHR
//SYSUT2 DD DSN=VICY.MQ1O.BSDS02,DISP=SHR
//SYSUT3 DD DSN=VICY.MQ1O.NEW.BSDS01,DISP=OLD
//SYSUT4 DD DSN=VICY.MQ1O.NEW.BSDS02,DISP=OLD

Externals – BSDS conversion utility
(CSQJUCNV)
● Typical output
CSQJ445I CSQJUCNV BSDS CONVERSION UTILITY - 2014-06-04
15:02:48
CSQU526I CSQJUCNV Connected to DB2 DB4A
CSQU528I CSQJUCNV Disconnected from DB2 DB4A
CSQJ200I CSQJUCNV UTILITY PROCESSING COMPLETED
SUCCESSFULLY

37
Log RBA – Using a 8-byte log RBA
• Queue manager issues message CSQJ034I during start-up
to indicate whether a 6 or 8-byte log RBA is in use
–CSQJ034I csect-name END OF LOG RBA
RANGE IS max-rba
• The print log map utility (CSQJU004) displays the BSDS
version
• The messages issued to warn that the queue manager is
approaching the end of the log RBA range remain the
same, but the thresholds have been updated

New message CSQJ034I on QMgr
startup
 Issued during QMgr startup
 Indicates whether QMgr is running in 6 or 8 byte RBA mode
- 0000FFFFFFFFFFFF – 6 byte RBA mode
11.25.05 STC05120 CSQJ127I !MQ4E SYSTEM TIME STAMP FOR BSDS=2014-04-02
11:19:18.70
11.25.05 STC05120 CSQJ001I !MQ4E CURRENT COPY 1 ACTIVE LOG DATA SET IS 280
280 DSNAME=VICY.MQ4E.LOGCOPY1.DS04, STARTRBA=00000000038F4000
280 ENDRBA=0000000003B0FFFF
11.25.05 STC05120 CSQJ099I !MQ4E LOG RECORDING TO COMMENCE WITH 281
281 STARTRBA=00000000039AF000
11.25.05 STC05120 CSQJ034I !MQ4E CSQJW007 END OF LOG RBA RANGE IS
0000FFFFFFFFFFFF
- FFFFFFFFFFFFFFFF – 8 byte RBA mode
22.57.53 STC13100 CSQJ001I !MQ08 CURRENT COPY 2 ACTIVE LOG DATA SET IS 810
810 DSNAME=VICY.MQ08.LOGCOPY2.DS01, STARTRBA=0000000002760000
810 ENDRBA=0000000003B0FFFF
22.57.53 STC13100 CSQJ099I !MQ08 LOG RECORDING TO COMMENCE WITH 811
811 STARTRBA=0000000002AA8000
22.57.53 STC13100 CSQJ034I !MQ08 CSQJW007 END OF LOG RBA RANGE IS
FFFFFFFFFFFFFFFF

CHINIT SMF

43
Chinit SMF: The Problem
 Prior to MQ v8.0 no SMF data for Chinit address
space or channel activity
 Many customers have had to create their own
‘monitoring’ jobs with periodic DISPLAY CHSTATUS
 Difficult to manage historical data, perform
capacity planning and investigate performance
issues

45
Chinit SMF: The Solution
 Additional SMF data for CHINIT address space and
channel activity to enable:
- Monitoring
- Capacity planning
- Tuning
 Separate controls from queue manager SMF
allows 'opt in'
 Updated MP1B SupportPac formats the data and
introduces rule based reporting

!MQ08 START TRACE(STAT) CLASS(4)
CSQW130I !MQ08 'STAT' TRACE STARTED, ASSIGNED TRACE NUMBER 05
CSQ9022I !MQ08 CSQWVCM1 ' START TRACE' NORMAL COMPLETION
!MQ08 START TRACE(ACCTG) CLASS(4)
CSQW130I !MQ08 ‘ACCTG' TRACE STARTED, ASSIGNED TRACE NUMBER 06
CSQ9022I !MQ08 CSQWVCM1 ' START TRACE' NORMAL COMPLETION
!MQ08 DISPLAY TRACE(*)
CSQW127I !MQ08 CURRENT TRACE ACTIVITY IS -
TNO TYPE CLASS DEST USERID RMID
02 STAT 01 SMF * *
05 STAT 04 SMF * *
06 ACCTG 04 SMF * *
END OF TRACE REPORT
47
CHINIT SMF controls
• You can start channel initiator statistics (STAT) trace by:
• You can start channel initiator accounting (ACCTG) trace by:
• You can display trace by:
•ALTER and STOP TRACE commands have also been updated

49
Chinit SMF: Controls
 SMF data collected on SMF interval of QMGR
- Can be same as z/OS SMF interval
 Chinit STAT trace allows high level view of activity
 Chinit ACCTG trace allows detailed view at channel level
– STATCHL attribute on queue manager to control system wide setting
– STATCHL attribute on channel to control granularity
– STATACLS attribute on the queue manager controls automatically
defined cluster sender channels.
– Data collected is a superset of that collected on distributed with
STATCHL message

51
CHINIT statistics summary
MV45,MQ20,2014/04/08,20:43:57,VRM:800,
From 2014/04/08,20:41:54.984681 to 2014/04/08,20:43:57.237939
duration 122.253258 seconds
Number of current channels..................... 20
Number of active channels .... ................ 20
MAXCHL. Max allowed current channels........... 602
ACTCHL. Max allowed active channels............ 602
TCPCHL. Max allowed TCP/IP channels............ 602
LU62CHL. Max allowed LU62 channels............. 602
Storage used by Chinit......................... 22MB

53
Dispatcher statistics
MV45,MQ20,2014/04/08,20:43:57,VRM:800,
From 2014/04/08,20:41:54.984681 to 2014/04/08,20:43:57.237939
Task,Type,Requests,Busy %, CPU used, CPU %,"avg CPU","avg ET"
, , , , Seconds, , uSeconds,uSeconds
0,DISP, 46, 0.0, 0.000000, 0.0, 12, 14
1,DISP, 168912, 1.4, 0.028218, 0.0, 8, 10
2,DISP, 168656, 1.3, 0.025142, 0.0, 7, 10
3,DISP, 0, 0.0, 0.000000, 0.0, 0, 0
4,DISP, 0, 0.0, 0.000000, 0.0, 0, 0
Summ,DISP, 337614, 0.6, 0.053360, 0.0, 9, 10
0,DISP, number of channels on this TCB, 0
Summ,DISP, number of channels on all TCBs, 20

55
Adapter statistics
MV45,MQ20,2014/04/08,20:43:57,VRM:800,
From 2014/04/08,20:41:54.984681 to 2014/04/08,20:43:57.237939
Task,Type,Requests,Busy %, CPU used, CPU %,"avg CPU","avg ET"
, , , , Seconds, , uSeconds,uSeconds
0,ADAP, 127599, 16.5, 0.953615, 0.8, 7, 158
1,ADAP, 46790, 7.6, 0.309678, 0.3, 7, 199
2,ADAP, 13702, 3.2, 0.065380, 0.1, 5, 284
3,ADAP, 2909, 0.7, 0.029541, 0.0, 10, 279
4,ADAP, 395, 0.1, 0.003179, 0.0, 8, 392
5,ADAP, 37, 0.0, 0.000241, 0.0, 7, 149
6,ADAP, 10, 0.0, 0.000175, 0.0, 17, 111
7,ADAP, 0, 0.0, 0.000000, 0.0, 0, 0
Summ,ADAP, 191442, 3.5, 1.361809, 0.1, 7, 179

57
Channel accounting data
• Accounting (ACCTG) trace CLASS(4) enables collection of
individual channel information
• Channels can be included/excluded from collection
– Channel STATCHL attribute for predefined channels
– Queue manager STATACLS attribute for auto-defined cluster channels
• Controls all auto-defined channels, not individually
– Queue manager STATCHL attribute for client channels
• Generally low cost as most channels are long lived
• Consider if you have a lot of short lived client connections
• Gives detailed information for each channel
– What has the channel done?
– What are my busy channels?

58
Channel information (1)
127.0.0.1 MQ89_1 Connection name 127.0.0.1
127.0.0.1 MQ89_1 Remote qmgr/app MQ89
127.0.0.1 MQ89_1 Channel disp PRIVATE
127.0.0.1 MQ89_1 Channel type SENDER
127.0.0.1 MQ89_1 Channel status RUNNING
127.0.0.1 MQ89_1 Channel STATCHL HIGH
127.0.0.1 MQ89_1 Channel started date & time 2014/04/08,19:41:48
127.0.0.1 MQ89_1 Channel stopped time
127.0.0.1 MQ89_1 Channel status collect time 2014/04/08,19:43:57
127.0.0.1 MQ89_1 Last msg time 2014/04/08,19:43:52
127.0.0.1 MQ89_1 Active for 122 seconds
127.0.0.1 MQ89_1 Batch size 50
127.0.0.1 MQ89_1 Messages/batch 38.9
127.0.0.1 MQ89_1 Number of messages 2,998
127.0.0.1 MQ89_1 Number of persistent messages 1,506
127.0.0.1 MQ89_1 Number of batches 77
127.0.0.1 MQ89_1 Number of full batches 42
127.0.0.1 MQ89_1 Number of partial batches 35
127.0.0.1 MQ89_1 Buffers sent 3,319
127.0.0.1 MQ89_1 Buffers received 109
127.0.0.1 MQ89_1 Xmitq empty count 13

59
Channel information (2)
127.0.0.1 MQ89_1 Message data 17,198,653 16 MB
127.0.0.1 MQ89_1 Persistent message data 4,251,780 4 MB
127.0.0.1 MQ89_1 Non persistent message data 12,946,873 12 MB
127.0.0.1 MQ89_1 Total bytes sent 17,200,221 16 MB
127.0.0.1 MQ89_1 Total bytes received 3,052 2 KB
127.0.0.1 MQ89_1 Bytes received/Batch 39 39 B
127.0.0.1 MQ89_1 Bytes sent/Batch 223,379 218 KB
127.0.0.1 MQ89_1 Batches/Second 0
127.0.0.1 MQ89_1 Bytes received/message 1 1 B
127.0.0.1 MQ89_1 Bytes sent/message 5,737 5 KB
127.0.0.1 MQ89_1 Bytes received/second 25 25 B/sec
127.0.0.1 MQ89_1 Bytes sent/second 140,985 137 KB/sec
127.0.0.1 MQ89_1 Compression rate 0
127.0.0.1 MQ89_1 Exit time average 0 uSec
127.0.0.1 MQ89_1 DNS resolution time 0 uSec
127.0.0.1 MQ89_1 Net time average 312 uSec
127.0.0.1 MQ89_1 Net time min 43 uSec
127.0.0.1 MQ89_1 Net time max 4,998 uSec
127.0.0.1 MQ89_1 Net time max date&time 2014/04/08,19:43:52

SCM Storage (Flash)

CF Flash: Scenarios Planned Emergency
Flash
CFStruct
Storage
90%
80%
70%
offload > 0KB
offload > 4KB
Offload > 32KB
SMDS
CFSTRUCT OFFLOAD rules
cause progressively smaller
messages to be written to
SMDS as the structure starts
to fill.
Once 90% threshold is
reached, the queue manager
stores the minimum data per
message (reference message)
to squeeze as many message
references as possible into the
remaining CF storage.
Once at 90% threshold, CF
Flash pre-staging algorithm
also starts to move reference
messages for new messages
arriving into the CF structure
into SCM (assume msgs are
off the same priority). Older
messages, which are likely to
be got first are kept in the
faster CF storage.
REF REF
REF REF
REF REF
Note: Assume all msgs < 63KB

CF Flash: Scenarios Maximum Speed
Flash
CFStruct
90%
80%
70%
Offload = 64KB => disabled
SMDS
We want to keep high performance
messages in the CF for most rapid
access.
CFSTRUCT OFFLOAD are configured
with special value '64k' to turn them off.
Once 90% threshold is reached, the CF
Flash algorithm starts moving new
messages to flash storage, keeping the
faster 'real' storage for messages most
likely to be gotten next.
As messages are got and deleted, the
CF flash algorithm attempts to pre-stage
the next messages from flash into the
CFSTRUCT so they are rapidly
available for MQGET.
In this scenario the flash storage acts
like an extension to 'real' CFSTRUCT
storage. However it will be consumed
more rapidly since all message data is
stored in it. Though, you could define a
threshold to offload >16KB messages to
SMDS if the CF structure is say 40%
full. This would mean that only
messages <=16KB ever get moved to
flash storage.
Note: Assume all msgs < 63KB
MSGMSG
MSGMSG

SCM Storage (Flash) – Comparison
Scenario Msg Size Total Msgs # in 'real' SMDS space # in f2la0s0h GB (Aliumgimt 3e0nGteBd)
NNoo FSlMasDhS 1kB 3M 3M
4kB 900,000 900,000
16kB 250,000 250,000
SMDNSo Flash 1kB 3.2M 3.2M 800MB
4kB 1.8M 1.8M 5GB
16kB 1.3M 1.3M 20GB
“EmSercgeennacryio” 1kB 190M 2M 270GB 190M 30GB
4kB 190M 600,000 850GB 190M 30GB
16kB 190M 150,000 3TB 190M 30GB
“SpeSecde”nario 1kB 150M 2M 150M 26GB
4kB 48M 600,000 48M 8GB

CFLEVEL(4) using 8KB Messages
1
4
7
C F L E V E L ( 4 ) 8 K M e s s a g e s - X M I T Q u e u e d e p t h
N o S C M a v a i l a b l e S C M a v a i l a b l e D e p t h t h a t S C M u s e d
1 0
1 3
1 6
1 9
2 2
2 5
2 8
3 1
3 4
3 7
4 0
4 3
4 6
4 9
5 2
5 5
5 8
6 1
6 4
6 7
7 0
7 3
7 6
7 9
8 2
8 5
8 8
9 1
9 4
1 0 0 0 0 0 0
9 0 0 0 0 0
8 0 0 0 0 0
7 0 0 0 0 0
6 0 0 0 0 0
5 0 0 0 0 0
4 0 0 0 0 0
3 0 0 0 0 0
2 0 0 0 0 0
1 0 0 0 0 0
0
T i m e ( s e c o n d s )
Q u e u e D e p t h
 Saw-tooth effect occurs when capture task goes into retry mode due to “storage medium full” reason code.
 Even with these 5 second pauses, the non-SCM capable workload completes in 90% of the time of the SCM
capable workload.
 Cost of workload in MVS differs by less than 2%.
 Get rate once the capture task has completed:
No SCM: 21100 messages / second ~ 164MB/sec
SCM: 19000 messages / second ~ 148MB/sec

Additional z/OS Items

73
64-bit application support
• Added 64-bit support for the C language
– LP64 compiler option
– Supported by cmqc.h
– No 64-bit COBOL
• Restricted environments
– BATCH, TSO, USS
– CICS and IMS do not support 64-bit applications
–WebSphere Application Server is already 64-bit
• Must use sidedecks & DLLs, not stubs:
– CSQBMQ2X (uncoordinated batch & USS)
– CSQBRR2X (RRS coordinated, srrcmit())
– CSQBRI2X (RRS coordinated, MQCMIT)

75
Client Attachment Feature
•In version 8 the Client Attachment Feature has
been removed
–Entitlement to connect clients is now included by default
•Client Attachment Feature also now non-chargeable
on MQ V701 and V710
–APAR PI13429 (PTF UI16087 and UI16088) available to
enable this capability without installing CAF
•Use CHLAUTH rules to restrict access if you’ve not
previously used clients

77
Multiple cluster transmission queues
• Support for multiple transmission queues in a cluster
– Separation of message traffic
– Management of messages
– System monitoring
– Implemented in version 7.5 on distributed platforms
• Externals
– DEFCLXQ queue manager attribute
• Default transmission queue setting for cluster-sender channels
• Use SYSTEM.CLUSTER.TRANSMIT.QUEUE or create a permanent-dynamic
transmission queue per cluster-sender channel called
SYSTEM.CLUSTER.TRANSMIT.<channel name>
– CLCHNAME queue attribute
• Set on a manually defined transmission queue
• Generic name for channels that should use it
– DEFINE QLOCAL(CLUSTER.XMITQ1) USAGE(XMITQ)
CLCHNAME(‘AAA.*’) …
• Most specific match is used by a channel

79
Multiple cluster transmission queues –
Switching
• A channel switches transmission queue in one of two ways:
– Automatically when the channel next starts
• Changes do not take effect while a channel is running
– Manually by CSQUTIL using the SWITCH CHANNEL function
• This is the equivalent of runswchl on distributed platforms
• Switching sequence
– Channel starts and resolves in-doubt status
– Channel initiates switch
– Channel switches to get messages from new transmission queue
• New messages continue to be put to the old transmission queue
– Queue manager starts moving messages for the channel from the old
transmission queue to the new transmission queue
– Switch completes when no committed or uncommitted messages for the
channel remain on the old transmission queue
• New messages now put to the new transmission queue

Multiple cluster transmission queues – Admin
• Version 8 new function must be enabled
• Various console messages output during switching process
•DISPLAY CHSTATUS and DISPLAY CLUSQMGR can
be used to view the transmission queue a cluster-sender
channel is using
•CSQUTIL reports the following for each cluster-sender
channel
–Transmission queue currently in use
–Pending / in-progress switch information
81

83
Additional z/OS items
• Message suppression
– CSQ6SYSP / SET SYSTEM property EXCLMSG
– Formalizes service parm to suppress Client channel start/stop messages
– Extended to be generalized
• Applicable for most MSTR and CHIN messages
• DNS reverse (ip address → host name) lookup
– Queue Manager attribute REVDNS(DISABLED/ENABLED)
- If DISABLED, prevents channel hangs if DNS infrastructure impacted
• But CHLAUTH rules that use hostnames are not matched
• Exploitation of zEDC compression hardware
– Channel attribute MSGCOMP(ZLIBFAST)
– Requires zEC12 GA2 + zEDC card
– Can yield higher throughput and reduced CPU for SSL channels

Additional z/OS items
 Integration of AMS capabilities into MQ for
z/OS
- Currently separately installable products
- Support of AMS with IMS Bridge
 Managed File Transfer has less reliance on USS

Migration and coexistence
 Backward migration and coexistence APAR for
MQ V701 and V710 is PI19721:
– PTF UI21966 for V701
– PTF UI21967 for V710
– FIXCAT IBM.Coexistence.MQ.V8R0M0
– SMP/E command:
REPORT MISSINGFIX ZONES(mqtgtzone)
FIXCAT(IBM.Coexistence.MQ.V8R0M0)

Questions?

For Additional Information
 IBM Training
 http://www.ibm.com/training
 IBM WebSphere
 http://www.ibm.com/software/websphere/
 http://www.ibm.com/software/products/ibm-mq
 IBM developerWorks
 http://www.ibm.com/developerworks/websphere
 https://www.ibm.com/developerworks/community/blogs/messaging
 WebSphere forums and community
 http://www.ibm.com/developerworks/websphere/community/

IBM MQ Sessions this week
10:30 - 12:00 13:15 - 14:15 14:30 - 15:30 16:00 - 17:00 17:15 - 18:15
Tue
sday
Opening General Session- IBM Digital
Experience and WebSphere Technical
University
Session A31: IBM
MQ CHLAUTH rules
– with MQ V8
updates
Speaker: Morag
Hughson
Room 02
Session A4: WebSphere
MQ for z/OS:
Performance and
Accounting
Speaker: Alexander Ross
Room 8
Session I26:
DataPower-MQ
Connectivity Deep Dive
(Theory)
Speaker: Robin Wiley
Room 27
Session Z1: WebSphere
MQ for z/OS V8: Latest
Features Deep Dive
Speaker: Damon Cross
Room 6
9:00 - 10:00 10:30 - 11:30 11:45 - 12:45 14:00 - 15:00 15:15 - 16:15 16:45 - 17:45
Wed
nes
day
Session Z5:
WebSphere MQ for
z/OS: Security
Speaker: Damon
Cross
Room 02
Session A21: What's
New in IBM
Messaging
Speaker: Morag
Hughson
Room 8
Session C7:
Messaging in the
Cloud with IBM MQ
Light and IBM
Bluemix
Speaker: Rob
Nicholson
Room 27
Session A17: Managing
work-loads, scaling and
availability with IBM MQ
clusters
Speaker: David Ware
Room 6
Lab IL5: DataPower-MQ Connectivity Deep Dive
(Hands-On)
Speaker: Robin Wiley
Room 7b
MQ for z/OS: The Inside
Story
Speaker: Damon Cross
Room 6
Thur
sday
Session A35: How to
Develop Responsive
Applications with IBM
MQ Light
Speaker: Rob
Nicholson
Room 27
Session A22: New
IBM MQ V8 Security
Features
Speaker: Morag
Hughson
Room 01
MQ for z/OS: Shared
Queues
Speaker: Alex Ross
Room 6
Session A18: Using
Publish /Subscribe with
IBM MQ
Speaker: David Ware
Room 27
Frid
ay
Lab AL6: Developing a First Application
with IBM WebSphere MQ Light
Speakers: Robert Nicholson, Alex Ross
Room 7b
Session A16: Using
IBM MQ Pub/Sub in

WebSphere Technical University and Digital
Experience Europe 2014
Your feedback is valuable - please complete your session
or lab evaluation!
Session number
Z1: IBM WebSphere MQ for z/OS Latest
Features Deep Dive
Provide your evaluations by:
Evaluation forms:
Fill out a form at the end of each session
Paper forms are located in each of the session or lab rooms
- Or –
Place the completed form in the tray as you exit the room
Complete the session survey on Event Connect Portal:
ibmeventconnect.eu/euxdx
Select Sessions, then Session Finder, and complete the
survey

Thank you!

IBM WebSphere MQ for z/OS V8 - Latest Features Deep Dive

More Related Content

Similar to IBM WebSphere MQ for z/OS V8 - Latest Features Deep Dive

Recently uploaded

IBM WebSphere MQ for z/OS V8 - Latest Features Deep Dive

Editor's Notes