Greg VanceIMS STSM June 11, 2015

IMS - Dynamic, Scalable, Efficient

IMS Tokyo RUG 2015

*

2

© Copyright IBM Corporation 2015. All rights reserv ed.U.S. Government Users Restricted Rights - Use, dupli cation or disclosure restricted by GSA ADP Schedule Contract with IBM Corp.

IBM’s statements regarding its plans, directions, a nd intent are subject to change or withdrawal without notice at IBM’s sole discreti on. Information regarding potential future products is intended to outline ou r general product direction and it should not be relied on in making a purchasing deci sion. The information mentioned regarding potential future products is no t a commitment, promise, or legal obligation to deliver any material, code or f unctionality. Information about potential future products may not be incorporated i nto any contract. The development, release, and timing of any future feat ures or functionality described for our products remains at our sole discretion.

IBM, the IBM logo, ibm.com, Information Management, IMS, CICS, DB2, WebSphere and z/OS are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with a trademark symbol (® or ™), these symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at “Copyright and trademark information” at www.ibm.com/legal/copytrade.shtml

Other company, product, or service names may be trademarks or service marks of others.

Disclaimer

Agenda

IMS Strategy continues to focus on enhancements that reduce outages and improve scalability

Taking advantage of these capabilities can improve your ability to provide service to your customers.

Introduction

DEDB Alter

Dynamic Refresh of P(WFI) regions

User Exit Enhancements

Other IMS 14 Enhancements

3

Data Cloud Engagement

Social. Mobile. Security. Empowering people with knowledge, enriching them through networks and changing expectations.

The emergence of cloud is transforming IT and business processes into digital services

Data is becoming the world’s new natural resource

There are three important shifts fundamentally changing IT

4

IMS Product Strategy is aligned with these shifts

� Core Capabilities– Reduce path length, contention,

I/O…

– Reduce planned outages

– More dynamic capabilities

� Data & Analytics

– Accelerate time to insight

� Cloud/Mobile

– Rapidly enable/control cloud &

mobile access to IMS resources

� Innovate and extend the

value of your IMS

investment

� Expand and empower the

IMS talent population

� Open interfaces & Java

� Modern tooling for

administrators,

developers and DBAs

Strategic Intent Investment

� Continue to deliver the

IMS Value Proposition

– Superior performance,

reliability, availability and

serviceability

– Minimize cost per transaction

5

DEDB Alter

IMS Technical Symposium 2015

Agenda

A sysprog can add capacity to DEDBs that have no SDEPs without taking an outage

Introduction

DEDB Alter

Dynamic Refresh of P(WFI) regions

User Exit Enhancements

Other IMS 14 Enhancements

7

Fast Path Data Entry Database (DEDB) Alter (IMS 13)

� Allows dynamic change to DEDB Area without unload/reload of the area – Alter physical attributes of DEDB Area: SIZE, UOW and/or ROOT parameters – Change the randomizer used for a DEDB Area– DBs with SDEPs not supported in IMS 13– Runs as a standard Fast Path IFP utility – Area remains online

� Requires the use of a two-stage randomizer allowing Areas to be processed individually

� Benefits– Improves DEDB Area availability by providing definitional modifications without taking

the Area offline – Improved management of DEDBs

DEDB Alter (ALTERAREA Function)

� ALTERAREA area_name– Changes UOW, SIZE, ROOT on DEDB DBD statement while DEDB online

� SIZEChange the CI size of an Area in a DEDB database

� UOW and ROOTChange the Root Addressable and Independent Overflow parts of DEDB area

– Changes Randomizer in RMNAME DEDB DBD statement while DEDB online� During DEDB Alter, read active DEDB Area with old randomizer

Migrate Active Area to Target area with new randomizer� After DEDB Alter, new randomizer replaces old randomizer

All Areas in DEDB database use new randomizerNew name must be 2-stage randomizerNew name must be different than original name active for DEDB area

– Only one active DEDB Area can change at a time– ALTERAREA does not support DEDB databases with SDEPs

� Can replace randomizer using REPLRAND function for DEDBs with SDEPs

9

10

BASEBASEBASEBASEBASEBASEBASEBASEDOVFDOVF

SMAPIOVFIOVFIOVF

900 UOWs 1000 CIs(100 UOWs)

8CIs

2CIs

RootAddressable

Part

IndependentOverflow

DEDB Alter Example (ALTERAREA)

DBD NAME=FPDEDB,ACCESS=DEDB,RMNAME=DEDBRANDAREA DD1=AREA1,...AREA DD1=AREA2,SIZE=4096,UOW=(10,2),ROOT=(1000,100)AREA DD1=AREA3,...

BASEBASEBASEBASEBASEBASEBASEBASEDOVFDOVF

SMAPIOVFIOVFIOVF

1800UOWs

2000 CIs(200 UOWs)

16CIs

4CIs

IndependentOverflow

DBD NAME=FPDEDB,ACCESS=DEDB,RMNAME=DEDBRAND

AREA DD1=AREA1,...

AREADD1=AREA2,SIZE=4096,UOW=(20,4),ROOT=(2000,

200)

AREA DD1=AREA3,...

BASEBASEBASEBASEBASEBASEBASEBASEDOVFDOVF

RootAddressable

Part

ALTERAREAUOW = (20,4)

ROOT = (2000,200)

Moving Forward

� IMS 13 support is the basis for future dynamic capabilities

� Support for all DEDBs

� Support for more DBD changes

� So that

A sysprog can add capacity to any DEDB without taking an outage

Dynamic Refresh of P(WFI) Regions

IMS Technical Symposium 2015

Agenda

A sysprog can refresh application programs in P(WFI) reqions without the need to stop and restart these regions.

Introduction

DEDB Alter

Dynamic Refresh of P(WFI) regions

User Exit Enhancements

64-Bit Capabilities

HALDB OSAM 8GB

TCO Improvements

Misc IMS 14 Enhancements13

14

IMS 14 Dynamic Refresh of (P)WFI Regions

� Today, when necessary to load a changed application program already scheduled and waiting for work in (P)WFI regions …

– The (P)WFI regions must be manually stopped in order to reload the application program

– It can be difficult to find all the regions to stop– It is time consuming to restart each affected region

� Introduce a new START(REFRESH) filter for the UPDATE PGM command enabling dynamic refresh for all instances of specified application program(s) running in (P)WFI dependent regions

– Reduces operational considerations for implementing a changed program– Dependent regions are not terminated, therefore remain available to process other work– Refreshed program gets rescheduled when a new work is available

IMS 14

START(REFRESH) Filter on UPDATE PGM Command

� UPDATE PGM NAME(pgmname) START(REFRESH)– Supported in both DRD and non-DRD environments– Support for multiple region types

� MPP (P)WFI regions with the program scheduled in them and program is not preloaded by the DFSMPLxx member

� JMP (P)WFI regions with the program scheduled in them� MPP, JMP and message driven BMP regions with the program scheduled in them

and transaction is defined as WFI=YES– Processed by all IMS systems that receive the command– The command processing is not coordinated among all IMS systems where the program

is scheduled

15

START(REFRESH) Filter on UPDATE PGM Command

� UPDATE PGM NAME(pgmname) START(REFRESH)– MPP and JMP regions remain available to schedule other work– BMP regions are terminated and must be restarted– MPP regions that have DBLDL=xx coded and if the program is in the dependent region

BLDL list� Program is removed from BLDL list when region is posted

– For JMP regions the JVM will be automatically recycled� DFSJVMMS (JVMOPMAS=), DFSJVMEV (ENVIRON=), DFSJVMAP configuration

members will be refreshed Next schedule of the program can obtain the refreshed copy of the JAVA program

� Changes to the Java classpath can be refreshed too !

16

Dynamic Refresh Restrictions

� UPDATE PGM with START(REFRESH) is not supported for the following region types– MPP regions where the application program is pre-loaded in the dependent region using

the DFSMPLxx PROCLIB member� These regions can be stopped via /STOP REGION command and the refreshed

program can be loaded with the next refresh– IFP regions– JBP regions– Not supported for non-message driven BMPs or JBP regions– Non-message-driven BMP regions

� UPDATE PGM with START(REFRESH) is not valid in an IMS DBCTL environment or on an XRF alternate

17

Dynamic Refresh Example

18

▪ UPDATE PGM NAME(pgmname) START(REFRESH) command example:

TSO SPOC Input:

TSO SPOC Output:

Note: There are numerous completion codes for the UPDATE PGM command

PgmName MbrName CC CCText RgnId RgnTypePGMA IMS1 0 1 MPPPGMA IMS1 0 3 MPPPGMA1 IMS1 221 Not supported for region type 2 IFPPGMA2 IMS1 223 Not PWFI region or WFI=Y TRAN 8 MPPPGMB IMS1 220 Program is not scheduledPGMMPL1 IMS1 224 Preloaded in DFSMPLxx list 9 MPPPGMNBMP1 IMS1 225 Non message driven BMP region 10 BMPPGMY IMS1 0 4 JMPPGX IMS1 10 Name not found

UPD PGM NAME(PGM*,PGX) START(REFRESH)

Dynamic Refresh of (P)WFI Regions

� IMS14 introduces a new START(REFRESH) filter for the UPDATE PGM command enabling dynamic refresh for all instances of the application program that are running in (P)WFI dependent regions

� Benefits– Reduces operational considerations for implementing a program change– Dependent regions are not terminated and remain available for other work– Refreshed program gets rescheduled when a new work is available

19

User Exit Enhancements

IMS Technical Symposium 2015

Agenda

A sysprog can refresh IMS system user exits without taking an IMS outageIntroduction

DEDB Alter

Dynamic Refresh of P(WFI) regions

User Exit Enhancements

Other IMS 14 Enhancements

21

Refreshable User Exits

� Refresh user exits without an IMS system outage– REFRESH USEREXIT command

� IMS 13 adds following exit “types” to support a list of exit names– AOIE Type-2 Automated Operator User Exit – BSEX Build Security Environment Exit (DFSBSEX0) – ICQSEVNT IMS CQS Event user exit– ICQSSTEV IMS CQS Structure Event user exit– IMSMON IMS Monitor User Exit – INITTERM Initialization/Termination user exit– LOGEDIT Log Edit Exit (DFSFLGE0)– LOGWRT Log Write Exit (DFSFLGX0)– NDMX Non-Discardable Message Exit (DFSNDMX0)– PPUE Partner Product user exit (DFSPPUE0)– OTMAIOED OTMA Input/Output Edit Exit (DFSYIOE0)– OTMARTUX OTMA Resume Tpipe Security Exit (DFSYRTUX) – OTMAYPRX OTMA Destination Resolution Exit (DFSYPRX0)– RASE Resource Access Security Exit (DFSRAS00)– RESTART Restart user exit

� Benefits– Improves availability – Simplifies user exit management

IMS 14

IMS 14

New Exits Enabled for Enhanced Services � Type-2 Automated Operator User Exit (AOIE - DFSAOE00)

– Enabled for enhanced use exit services� QUERY / REFRESH AOI User Exit routines

QUERY USEREXIT TYPE(AOIE)REFRESH USEREXIT TYPE(AOIE)

� Supports multiple user exit routines for AOIE exit point

� New IMS Monitor User Exit (IMSMON)– Enabled for enhanced user exit services

� QUERY / REFRESH USEREXIT IMS Monitor User Exit routinesQUERY USEREXIT TYPE(IMSMON)REFRESH USEREXIT TYPE(IMSMON)

� Supports multiple user exit routines for IMSMON exit point– Architected IMS Monitor user exit interface

� A clearly differentiated programming interface (CDPI) to allow capture of data available to the IMS Monitor

23

Type-2 Automated Operator User Exit (AOIE)

� Define exit routines in the EXITDEF parameter with TYPE=AOIE in the USER_EXITS section of the DFSDFxxx proclib member

– Can use standalone module named DFSAOE00 in STEPLIB instead

� Called continuously for IMS system messages that are …– destined for master terminal– for type-1 commands that are entered from terminals or from APPC or OTMA

connections– for the responses to type-1 commands entered from terminals or APPC or OTMA

connections

� The exit routine is also called for internal commands and commands from ICMD calls, but not for the responses24

<SECTION=USER_EXITS>

EXITDEF=(TYPE=AOIE, EXITS=(AOEXIT1,AOEXIT2))

IMS Monitor User Exit

� Define exit routines in the EXITDEF parameter with TYPE=IMSMON in the USER_EXITS section of the DFSDFxxx proclib member

� Exit(s) called at IMS, initialization, termination and Monitor log record creation

� Exit(s) called with INIT call function code after successful REFRESH USEREXIT

� When the IMS Monitor is active, the IMS Monitor is called first, then the exit(s)

� There are no changes to IMS Monitor log records or processing

� Sample exit routine provided in SDFSSMPL(DFSMONX0)– Demonstrates how to register interests for SLOG codes during the INIT call– Does not take any action

25

<SECTION=USER_EXITS>

EXITDEF=(TYPE=IMSMON, EXITS=(MONEXIT1,MONEXIT2))

User Exit Enhancements

� IMS 14 extends the enhanced user exit support – Enables exit query/refresh and ability to call multiple routines for two user exits:

� AOIE (DFSAOE00)� New IMSMON� Provides a documented user exit interface to the IMS Monitor

� Benefits– IMS availability is increased by making additional user exits refreshable– Provides access to the IMS Monitor data without the need to modify IMS code

26

Other IMS 14 Enhancements

IMS Technical Symposium 2015

Agenda

IMS 14 provides several enhancements that can reduce outages, allow you to increase capacity, and simplify operations.

FP utility 64-bit support

HALDB OSAM 8G support

IMS Connect Enhancements

OTAM Enhancements

External subsystem support enhancements

Introduction

DEDB Alter

Dynamic Refresh of P(WFI) regions

User Exit Enhancements

Other IMS 14 Enhancements

28

Growing your Fast Path Data Entry Databases

29

� IMS 11 introduced a 64-bit buffer manager to take the guess work out of defining Data Entry Database (DEDB) buffer pools

� IMS manages the subpools automatically

� Multiple subpool sizes are used based on the database

� High Speed Reorganization (HSRE) and High Speed Sequential Processing (HSSP) buffers can now use 64-bit storage

� Why this is important

– Reduces need for 31-bit common storage (ECSA)

– Allows the HSSP and HSRE utilities to run against UOWs with a very large number of segments

Growing your Full Function Databases � High Availability Large Database (HALDB) enables independent partitions for

your full function database allowing growth greater than 40 terabytes– Supports OSAM and VSAM

� Use HALDB Online Reorganization to reorganize partitions without an application outage

� Allow database growth by supporting 8GB OSAM for HALDBs – You can choose either 8GB OSAM HALDB or the use of Online

Reorganization for each database

� Reduce 24-bit storage usage for OSAM Database control blocks– OSAM DEB extension moved into 31-bit storage

� Why this is important– IMS 14 provides increased scalability for OSAM databases

30

Enabling Cascaded Transactions� Distributed global transactions can combine IMS transactions with other workload

requests (WAS, CICS, DB2 requests) into one unit of recovery

� Resource Recovery Services (RRS) coordinates the syncpoint processing for global transactions

� Allow a Global Transaction using the IMS TM Resource Adapter to be spread across IMS Connect and an IMS Control region that reside on different LPARs

– For those transactions that require 2-phase commit (SYNCLVL=SYNCPT)

� Why this is important– Increases flexibility and workload balancing of IMS across LPARs – Increases availability by allowing IMS Connect to route to another LPAR

31

Distributed WAS with

IMS Resource Adapter

IMSConnect

z/OS z/OS

IMS

Improving scalability for OTMA Transaction Pipes � Allow multiple active Resume TPIPEs for a single TPIPE

� TPIPEs sending CM0 output require ACK before next message can be sent � Multiple IMS Applications issuing ICAL to a single TPIPE

� Why this is important– Potential reduction in application wait time for synchronous callout to an external

server – Potential throughput improvement for asynchronous message delivery– Provide additional Resume TPIPEs for failover; throughput or workload balancing

32

Client1

Resume TPIPE ALTCID=ICONTP1

Client2

Resume TPIPE ALTCID=ICONTP1

Client3

Resume TPIPE ALTCID=ICONTP1

TPIPE= ICONTP1Message1Message2Message3

Managing change in IMS Connect � Commands allow dynamic change for IMS Connect configuration

� Allow additional commands to dynamically manage IMS Connect resources – Introduce DELETE PORT and DATASTORE commands– Enhance UPDATE PORT and DATASTORE commands to allow changes to

selected attributes– Introduce CREATE and DELETE IMSPLEX commands– Convert selected IMS Connect commands to complete synchronously

� Command response would reflect actual command result

� Why this is important– Availability is increased by allowing dynamic creates, deletes and updates

for selected resources– Usability is increased by providing a synchronous command response

33

Defining External Subsystems

34

� External Subsystem Attach Facility (ESAF) allows IMS applications to communicate with other subsystems

� Provide options to define WebSphere MQ and WebSphereOptimized Local Adapter subsystem in addition to DB2

� New subsystem type (SST=) values

� SST = DB2 | MQ | WOLA

� Positional SSM definitions assume SST=DB2

� Commands and log records changed to show SST value for subsystem

� /DIS SUBSYS

� /DIS OASN SUBSYS

� Why this is important

– Improves usability

35

FDBR Resolve In-Doubt Notification Exit – DFSFIDN0� Fast Database Recovery (FDBR) regions monitor an IMS subsystem and can

automatically recover database resources if that IMS fails, reducing the time a data sharing subsystem must wait for locks

� Simplify the resolution of in-doubt work for DB2 when Fast Database Recovery releases IMS – New sample DFSFIDN0 user exit that can be used without having to write

custom code to enable automation to resolve in-doubt work– Issues a message for each in-doubt ESS UOR with the subsystem

information – Automation can monitor the messages and act upon the in-doubt UOR

� Why this is important– Improves usability when using both FDBR and DB2 z/OS– Can reduce the time DB2 holds locks during an IMS abend situation

Thank You!

IMS Technical Symposium 2015