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 [email protected]

© 2014 IBM Corporation © 2014 IBM Corporation

Please NoteIBM’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…

64-bit Buffer Pools

Z1: IBM WebSphere MQ for z/OS Latest Features Deep Dive


Buffer pools – BackgroundQ1

Page Set

4KB Page

Page Set

Q2


Buffer pools – Using only 31-bit storage

DATA

CODEDATA

2 GB BAR

Queue Manager Address Space

Buffer Pool Buffer Pool Buffer Pool Buffer Pool Max 1.6GB for Buffer Pools

16 EB


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


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


Buffer pools – Using 64-bit storage

DATA

CODEDATA

2 GB BAR

Queue Manager Address Space

Buffer Pool

16 EB

Buffer Pool Buffer Pool Buffer Pool

Buffer Pool


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


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


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 88CSQ9022I !MQ21 CSQPALTB ' ALTER BUFFPOOL' NORMAL COMPLETIONCSQP023I !MQ21 Request completed for buffer pool 88, now has 12000 buffersCSQP054I !MQ21 Buffer pool 88 is now located above the bar


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


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



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.


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'


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)


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 15:02:48CSQU526I CSQJUCNV Connected to DB2 DB4A CSQU528I CSQJUCNV Disconnected from DB2 DB4A CSQJ200I CSQJUCNV UTILITY PROCESSING COMPLETED SUCCESSFULLY


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

− FFFFFFFFFFFFFFFF – 8 byte RBA mode

11.25.05 STC05120 CSQJ127I !MQ4E SYSTEM TIME STAMP FOR BSDS=2014-04-02 11:19:18.7011.25.05 STC05120 CSQJ001I !MQ4E CURRENT COPY 1 ACTIVE LOG DATA SET IS 280 280 DSNAME=VICY.MQ4E.LOGCOPY1.DS04, STARTRBA=00000000038F4000 280 ENDRBA=0000000003B0FFFF11.25.05 STC05120 CSQJ099I !MQ4E LOG RECORDING TO COMMENCE WITH 281 281 STARTRBA=00000000039AF00011.25.05 STC05120 CSQJ034I !MQ4E CSQJW007 END OF LOG RBA RANGE IS 0000FFFFFFFFFFFF

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=0000000003B0FFFF22.57.53 STC13100 CSQJ099I !MQ08 LOG RECORDING TO COMMENCE WITH 811 811 STARTRBA=0000000002AA800022.57.53 STC13100 CSQJ034I !MQ08 CSQJW007 END OF LOG RBA RANGE IS FFFFFFFFFFFFFFFF

CHINIT SMF



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


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


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

!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


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


CHINIT statistics summaryMV45,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 secondsNumber 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


Dispatcher statisticsMV45,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 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, 0Summ,DISP, 337614, 0.6, 0.053360, 0.0, 9, 10 0,DISP, number of channels on this TCB, 0 1,DISP, number of channels on this TCB, 10 2,DISP, number of channels on this TCB, 10 3,DISP, number of channels on this TCB, 0 4,DISP, number of channels on this TCB, 0Summ,DISP, number of channels on all TCBs, 20


Adapter statisticsMV45,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 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, 0Summ,ADAP, 191442, 3.5, 1.361809, 0.1, 7, 179


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?


Channel information (1)127.0.0.1 MQ89_1 Connection name 127.0.0.1127.0.0.1 MQ89_1 Remote qmgr/app MQ89127.0.0.1 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 time127.0.0.1 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


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)


© 2014 IBM Corporation

Flash

CFStruct

CF Flash: Scenarios Planned Emergency Storage

70%

80%

90%

Offload > 32KB

offload > 0KB

offload > 4KB

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


Flash

CFStruct

CF Flash: Scenarios Maximum Speed

70%

80%

90%Offload = 64KB => disabled

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.

Offload = 64KB => disabled

Note: Assume all msgs < 63KB

MSGMSG

MSGMSG


SCM Storage (Flash) – Comparison

Scenario Msg Size Total Msgs # in 'real' SMDS space # in 200 GB flash

Augmented(limit 30GB)

No SMDSNo Flash

1kB 3M 3M

4kB 900,000 900,000

16kB 250,000 250,000

SMDSNo Flash

1kB 3.2M 3.2M 800MB

4kB 1.8M 1.8M 5GB

16kB 1.3M 1.3M 20GB

“Emergency”Scenario

1kB 190M 2M 270GB 190M 30GB

4kB 190M 600,000 850GB 190M 30GB

16kB 190M 150,000 3TB 190M 30GB

“Speed”Scenario

1kB 150M 2M 150M 26GB

4kB 48M 600,000 48M 8GB

16kB 12M 150,000 12M 2GB


CFLEVEL(4) using 8KB Messages

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/secSCM: 19000 messages / second ~ 148MB/sec

1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94

0

1 0 0 0 0 0

2 0 0 0 0 0

3 0 0 0 0 0

4 0 0 0 0 0

5 0 0 0 0 0

6 0 0 0 0 0

7 0 0 0 0 0

8 0 0 0 0 0

9 0 0 0 0 0

1 0 0 0 0 0 0

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

T i m e ( s e c o n d s )

Qu

eu

e D

ep

th

Additional z/OS Items



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)


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


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


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


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


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

Additional z/OS items


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)

Migration and coexistence

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 week10:30 - 12:00 13:15 - 14:15 14:30 - 15:30 16:00 - 17:00 17:15 - 18:15

Tuesday

Opening General Session- IBM Digital Experience and WebSphere Technical University

Session A31: IBM MQ CHLAUTH rules – with MQ V8 updates

Speaker: Morag HughsonRoom 02

Session A4: WebSphere MQ for z/OS: Performance and Accounting

Speaker: Alexander RossRoom 8

Session I26: DataPower-MQ Connectivity Deep Dive (Theory)

Speaker: Robin WileyRoom 27

Session Z1: WebSphere MQ for z/OS V8: Latest Features Deep Dive

Speaker: Damon CrossRoom 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

Wednesday

Session Z5: WebSphere MQ for z/OS: Security


Session A21: What's New in IBM Messaging


Session C7: Messaging in the Cloud with IBM MQ Light and IBM Bluemix

Speaker: Rob NicholsonRoom 27

Session A17: Managing work-loads, scaling and availability with IBM MQ clusters

Speaker: David WareRoom 6

Lab IL5: DataPower-MQ Connectivity Deep Dive (Hands-On)

Speaker: Robin WileyRoom 7b

Session A9: WebSphere MQ for z/OS: The Inside Story


Thursday

Session A35: How to Develop Responsive Applications with IBM MQ Light

Speaker: Rob NicholsonRoom 27

Session A22: New IBM MQ V8 Security Features


Session A3: WebSphere MQ for z/OS: Shared Queues

Speaker: Alex RossRoom 6

Session A18: Using Publish /Subscribe with IBM MQ


Friday

Lab AL6: Developing a First Application with IBM WebSphere MQ Light

Speakers: Robert Nicholson, Alex RossRoom 7b

Session A16: Using IBM MQ Pub/Sub in an MQ network


WebSphere Technical University and Digital Experience Europe 2014 Your feedback is valuable - please complete your session

or lab evaluation!

Session numberZ1: 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

Place the completed form in the tray as you exit the room- Or –

Complete the session survey on Event Connect Portal: ibmeventconnect.eu/euxdx

Select Sessions, then Session Finder, and complete the survey

Z1: IBM WebSphere MQ for z/OSLatest Features Deep Dive

Thank you!

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

Technology

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