Minutes of the GSE UK IMS Working Group Meeting - South Bank, 9th April 1999

Chairman: Frank Fleming (Frank.Fleming @ barclays.co.uk) - 01565-613581

Secretary: Peter Armstrong (peter_armstrong @ bmc.com) - 01276-419835

Administration

Introduction

• Bit of a sparse turnout - Easter holidays? Welcome to all and a run through the agenda. Wanted to hijack someone on way to IMS/DB2 Tech Conference, but there was no particular single session we wanted. Alan had someone lined up, but she had to cancel as she is moving house.

• Maintenance strategy is being brought in to the agenda today - unfortunately Alan Lovett, who suggested the session has taken his wife to Paris for her birthday, but he has sent notes to Frank, which Frank will share with us later in the meeting.

Minutes

• Everybody had received. Minutes and agenda are out on Web (to be found at http://web.ukonline.co.uk/dougie.lawson/gse) and agenda is on DIAL.

Matters arising

• None.

New Members / attendees

• None.

Apologies

• Got lots of apologies - thank you, we need to know who is still interested.

• Standard Life, Cap Gemini, J&H Marsh & McLennan, EDS Derby, Guradian Insurance, AIB, Integris, EDS, IBM (Alison Coughtrie), Royal Sun Alliance, ITSA

Correspondence

• Proceedings from GSE UK National Conference in November. Flyers for OS/390 and Storage Systems Technical Conference - Madrid 3-7 May..

General

• Frank wandered round the Web and produced a meeting calendar for other GSE Working groups plus write-ups of their objectives for people to look at.

• DBCTL subgroup in March cancelled due to illness and agenda problems. Talk to Ian if you want to help / have ideas for topics. Next meeting is on 1^st July, which is a joint meeting in Manchester with us. Final meeting will be October 28^th in South Bank. We need ideas / volunteers for sessions. Neil Kenyon for Web?

IBM What's New - Alan "I hate computers" Cooper (EMEA Technical Sales, IBM UK)

• Alan has suffered PC hard disk failure - backup machine lost half its memory!

• CA09 - DBRC will become a five day workshop with hands-on bits. Alan went through a list of the offerings - contact Alan or Pete if you need anything else.

• IMS/DB2 Technical Conference is next week in Eurodisney - two major themes are Web access and Sharing. US Tech Conference is in August in San Diego. Frank - also look at SHARE in US.

• Running an e-IMS road show in Europe in June and July. Alan wants to know if there is enough interest to run one in the UK. Ian - run it back-to-back with GUIDE in Manchester. Designed for IMS people so that they can talk intelligently to the young lads with long hair and sandals who rabbit on about Java.

• Icing group has announced four new tools - details in the afternoon.

• IMS News - Barbara Frost sends out to an email distribution list. To register, simply send a note to imsnews@us.ibm.com.

• Some new and existing redbooks for you to look at - have a look at www.redbooks.ibm.com, and an article in the IBM Systems journal about IMS.

• A copy of the foils was distributed.

BMC What's New? (Peter Armstrong - BMC Software)

• Peter took us through a review of what is new with BMC Software and what the merger with Boole and Babbage means for customers.

• First he reviews several new IMS products, which BMC has brought out recently:

• Data Movement - BMC can move data from IMS, DB2 and VSAM to DB2, Oracle, Sybase, SQL Server etc. They can do this in two flavours - Bulk Data Move, e.g. a regular refresh or a starting point, and Change Data Move - capturing data on the source system and moving them in near real-time (latency ca. < 1 sec) to a target system. Products look after extract/capture, transform, transport and load/apply from common interface.
• Snapshot utilities - the ability to run utilities concurrent with online update. Copy and Check available for IMS and now able to exploit the underlying hardware as well. Commands, RECON updates, hardware intervention etc. are all automatic.
• System Wide Point of Consistency (POC) - BMC Recovery products enable you to recover to any point in time. This applies to single IMS, sharing, IMS and DB2, VSAM, OS data sets or any combination of the above. Great for disaster recovery and/or logical recovery (i.e. finger trouble!). This means you can run systems and recover them without having to do /DBRs all the time to establish recovery points.
• Deltaplex - a new version of DELTA has been developed which makes changes across all IMS systems in a plex. Product will check that change is possible in all systems and execute - if problems, don't apply change to any system. Lots of other nice new goodies as well.

• B&B

• The merger of BMC and B&B effectively means that BMC now have the instrumentation facility and Automation engine of Mainview available to them. This means that you can use the software to determine when things are going wrong, when utilities need to be run, when performance is suffering etc. and trigger the appropriate actions to correct the problem. This means that you have to spend less time and effort on monitoring the system and can run housekeeping when it is needed rather than when you think it might be a good idea.

• A copy of the foils was distributed.

Maintenance strategies at Barclays Bank (Frank Fleming)

• Frank described how they do IMS maintenance at Barclays. They use DELTA on their development systems, but not on the production systems (for financial reasons!). Frank took us through their methodology for regular maintenance, fixes for problems, program products etc. They try to stay 3 months behind current.

• They have ten or eleven scheduled outage slots per year (freeze in December).

• Dougie - biggest variation is how recent people remain. Some people are 3 months behind, others haven't installed any maintenance for years, and are now installing strings of PTFs for Y2K.

XXSERV (Dougie Lawson, IBM UK)

• Dougie demonstrated the UK S/390 software support part of XXSERV, e.g. showing us the status of current PMRs with the ability to update / request close etc. The service has only provided since March - the goal is for it to be a replacement for DIAL IBM. Try registering for it - it will let you know if you have fulfilled the contractual requirements (i.e. are you a bona-fide paid-up punter). http://www.europe.ibm.com/xxserv.html

• He also demonstrated searching closed APARs, open APARs etc. (do you put in keywords or does it search the whole document? Not sure), and the enhanced HOLDDATA system. The enhanced HOLDDATA system comes as from OS/390 1.3 and if you have been using SMARTdata for years, you have a change coming. It was introduced via APAR IR35282 for OS/390 V1R3 and V2R4, and it is integrated into OS/390 V2R5. I use ASAP on DIAL - this is a replacement for ASAP. http://service.software.ibm.com/390holddata.html

• Another new facility is SUF - Service Update Facility. Dougie still playing with this to see how it works and what it can deliver.

• As a finale he demonstrated a Web based Tic-Tac-Toe game running on IMS - obviously he has too much time on his hands!

• IMS does not support DBs in non-ICF catalogs (and they aren't supported beyond the end of this year anyway, so get off now!!!

Problems forum, hints and tips, contacts for lunch

Requirements - none.

Dougie - list of PTFs needed for Y2K for 4.9, 5.1 and 6.1 - Dougie will stick on his Website, or contact him direct.

• Name: Duncan Mead - Liverpoool Victoria Friendly Society

• Problem: Has BLS course been cancelled? Been told it will only be run at user sites. Anyone else interested?

• Contacts/Comments: Alan will check with Pete as he handles that course.

• Name: Chris Huffington - Friends Provident - 01306-653426

• Problem: Time required to implement MQ - hints & tips ?

• Contacts/Comments:

• Name: Reshma Chotalia - Bank of America

• Problem: Anybody experienced problems with IMS 6.1, PUT 9901?

• Contacts/Comments: Dougie - should be fine if you are not using shared message queues or fast path. Try contacting Robert Queen at Bank of Scotland

• Name: Nuzhat Farooqi - Lloyds TSB Group

• Problem: Changing maintenance strategy from CBPDO to SystemPaks - any experiences?

• Contacts/Comments: Dougie - built by non-IMS people, based on theory that you have never built system before and have no USERMODs. Long discussion on the issue. They have got a lot better, but they are still not right. They will only get better if folks fill in the "installer's comment form" to report errors not just in the docs but also in the generated JCL etc.

  LUNCH

  IMS Tools - Alan Cooper (IBM)

• Tools are written by e.g sub-contractors and are marketed by IBM.

• Alan Cooper ran through a review of the four new IMS Tools available from Icing. Several of these use a new facility called XCM, which ships data from one address space to another within one MVS.

• IMS Parallel Change Accumulate - front-end to standard Change Accum utility, which enables you to run Change Accums in parallel. The logs are all read by the first job, which passes any log records it doesn't need to the next job using XCM, which in turn passes on to the next one etc. For DBRC support, you have to use a REXX EXEC called ICBGENRX, which repeatedly calls DBRC to generate the JCL for each job in the job stream. Each Change Accum runs in a separate address space.

• IMS Parallel Reorganisation - Unload, Reload and Index Builder run in parallel. Not same products as HSSR, FRR - new Unload (IPR Unload) and Reload (IPR Reload). Other components are IPR Scan, IPR Index Builder (same as standard IIB) and IPR Prefix Resolution. These are faster than the standard IMS utilities and also faster than the DBT utilities. No API in this tool - still available in HSSR. DBRC must be switched off for part of this parallel reorg, because it is running load and unload in parallel.

• IMS Data Stream Tuner - IMS outbound datastream reduction by eliminating repeating characters.

• IMS ETO Support Tool - makes ETO easier to use, with dynamic front-end (which happens to be IMS).

MQ Series (Frank Fleming)

• Have been using MQ for some time and have one application in production.

• Discussion about security and where to implement it. Frank passes userid and password through to IMS, and he is not interested in what is happening beyond first MQ next to IMS.

• Graham Mason at EDS is about to implement a new MQ application, and there was a discussion on different implementation methods and how many MQ licences each one needs.

• There are different ways of getting into IMS from MQ. Fast Path EMH has to use commit mode 1 (= send then sync). IBM told Frank that OTMA MQ Bridge is transparent and they are right. Duncan has DBCTL and hence uses BMP route in.

• Frank then did his presentation on how to do it back to front, where they have connected OS/2 to RS/6000 to IMS to MQ to Tandem. They did this because the RS/6000s were already attached to IMS, which in turn was already attached to MQ and it saved them buying 2000+ RS/6000 MQ licences. But, how does IMS work out what LTERM to use to send - they decided to use a table of LTERMs. Frank also described how to get round problems of EBCDIC boxes talking to ASCII boxes. Dougie - beware how numbers are stored, sometimes it is easier to store as character which is at least recognisable at both ends.

• Why didn't you simply attach RS/6000 direct to Tandem via APPC - can't remember, but one reason is that MQ is strategic. Assured message delivery.

• Don't use the trigger monitor - to quote several people "YUCK".

• A copy of the foils was distributed.

AOB

• IMS Listerver - this is like a chat forum and will let you ask questions / look at other people's Q&As. The address is listserv@mizzou1.missouri.edu. If you send the following email, you will become a member of the list:

• SUBSCRIBE IMS-L your name

• That should be all that is included in the message text. You will receive a confirmation message. The subject can be anything.

• Please submit ideas for future meetings / please volunteer.

• Next meetings: 1^st July for full group and 28^th October for IMS group.