Db2 update day 2015 managing db2 with ibm db2 tools svenn aage

© 2015 IBM Corporation
zAnalytics
DB2 Update Day 2015 – March 23-27, 2015
1
DB2 Update Day 2015
Svenn-Aage Sønderskov
Systems Programmer
JN Data A/S, Denmark
sas@jndata.dk

zAnalytics
2
Managing DB2
with IBM DB2 Tools

zAnalytics
3
Agenda
Who is JN Data ?
Setting the scene – JN Data Consolidation
–How did we do it ?
Focus on User productivity
–Moving the ”goodies” into the product
Questions

zAnalytics
4
JN Data – Where are we located
Silkeborg
datacenter
Roskilde
Kompetencecenter

zAnalytics
5
Numbers for thought
We give our customers competitive power
through a safe, stable and efficient IT
platform
We give our customers competitive power
through a secure, stable and efficient IT
platform
• JN Data's customers have a market share of approximately 35% of the total
banking market in terms of balance-sheet total
• More than 200 banks use JN Data's infrastructure
• JN Data's customers have a market share of more than 50% of the total
mortgage-credit market
• +40,000 computers are connected to JN Data's network. This means that approx.
42% of the employees in the financial sector use systems which are operated by
JN Data
• JN Data administer approx. 10 Petabyte data.
• The operation and Infrastructure support the work of +2000 developers
• 4 Systemic risk Financial Service Institutes run on the infrastructure and so does
the Danish National bank (Clearing and RTGS systems)
• The Danish Financial Service Authority have concluded that JN Data operates one
of the absolute most important datacenters in the country.

zAnalytics
6
8 of 10 Service Prices are lower with the Reference
Group average and 2 are slightly higher

zAnalytics
7
8 of 10 Service Prices are within the “33% percentile”
On 8 out of 10 areas JN Data outperforms the best comparison companieser.
91,5 % af cost in scope is within the best third of the groups. JN Data’s official goal is 80%.

zAnalytics
8
Why Consolidation ?
Many vendors/products - Reduction needed
Part of cutting cost by 3. digit million DKK p.a.
We had to keep the same or better service/performance
We had to move many years of experience from one set of products to the chosen ones.
We had to do it with 1/3 of the previous resources due to manpower consolidation.

zAnalytics
9
The Job
Streamlining the DBA-tools Suite for Change, Edit, Catalog and Application Tuning.
We had to reduce the number of vendors from 3 to 1 and products accordingly.
We had to do it within 12 Months from decision to end of contract for the first product line.
We had to make a significant cost reduction for JN Data’s Partners.

zAnalytics
10
What did we do at Analyze time?
We had to be very fact-based
– Can the vendor solve the task both on a Short and Long Term basis ?
– Can the products do the job ?
– Can the products support the current and new releases of the base DB2 software ?
– Setting a multi-question and answer was the chosen approach.

zAnalytics
11
Decisions on the way
DB2 Change Management is a critical service due to constraints like 24 by 7, Data volumes
etc.
We have 2 implementations for doing Change Management.
– User-driven where the user request production turnover
– DBA driven, where the DBA’s do the production Turnover.
Both in house developed frameworks building on top of the vendor product lines – Kept for
future use
– We might have to Re-Choose
We chose to maintain both with the chosen product line as engine. This offers us the ability
to change to another product line without affecting our users.
We Chose to go for the Change Management Method

zAnalytics
12
We analyzed the Solutions
We evaluated 3 vendors solutions
– Could they support the current and future DB2 Features ?
– Could they be embedded in our home-grown solutions and be operated in Batch?
– Could they solve a given number of specified tasks – Defined by all participants on a
“NEED TO HAVE” / “NICE TO HAVE” basis?
– Could they support the new features of DB2 10 (that was the one we started at).
– We summarized what each single JN Data Partner would gain/loose from the
consolidation and what effort they would need to be put into the transformation.

zAnalytics
13
What if ?
If we did not consolidate
– Would we pay for 3 solutions on a big MIPS footprint (Money)?
– We would have 3 contracts instead of one
– We would have to Maintain 3 solutions
– If one productline did not support the new releases of DB2 – should we wait until all 3 did
before upgrading to the new release of DB2?
– We decided to do consolidation even if one of the other product lines should turn out to
be better.

zAnalytics
14
What did we do with Change ?
Change - Built the chosen products into our home-grown framework – This was the
time/resource consuming part.
– Seen from the user the task performed has to behave as before.
– DDL Files in – Change/validation Rules Objectt Comparison – Worklist out
– We feel that we are ready for future choices ☺
Catalog/Edit/Application Tuning
– This was the easy part.
– Consolidation on the chosen product-line, Education (New commands had to be
learned)
– Converted our homegrown commands from the previous solutions into the new toolset
to ensure productivity at the user-level and thus upgrading all our partners with all the
goodies.

zAnalytics
15
DB2 Administration Tool
Basically DB2 Administration Tool is a standard ISPF application, which like the others can
help administering DB2 on System z.
DB2 Administration Tool can navigate the catalog and do most of the tasks needed
The setup can be a little challenge when operating on many clients with different setups and
many DB2 members and Datasharing Groups therefore we have had to make some new
things:
– Init REXX – Initialisation of dialogues variables, Libraries etc
– We have had to make changes to the behaviour on Templates and some of the
dialogues (Tables, Packages, Primary Commands) to get our need for behaviour
supported

zAnalytics
16
Extentions implemented on CMDS
Commands
– <JN> JN Data Homegrown Commands
– DROPALL ISPF SELECT CMD(%JNADBDRA &CPANEL &ÅTABLE)
– OID ISPF SELECT CMD(%JNADBOID)
– REBALL ISPF SELECT CMD(%JNADBRBA &CPANEL &ÅTABLE)
– FREEALL ISPF SELECT CMD(%JNADBFRA &CPANEL &ÅTABLE)
– GETSORT ISPF SELECT CMD(%JNADBSRT)
– <IBM> System Commands
– QTAB ISPF SELECT CMD(%ADBEDIAG QUERYTAB)
– TOOLS ISPF SELECT CMD(%ADBEDIAG)
– SHOWPAN ISPF SELECT CMD(%ADBEDIAG SHOWPANEL)

zAnalytics
17
Sample Rexx
/* REXX sample REXX for the DROPALL function */
ARG CPANEL ÅTABLE
If Wordpos(CPANEL,'ADB21S ADB21T ADB21VV ADB21X') = 0 then do
Say 'DROPALL is not supported in this panel '
Say 'Operation Aborted '
Exit
end
Address ISPEXEC
"VGET (DB2N) SHARED"
if DB2N = 'DRDA' then do
Say 'DROPALL is only possible Locally'
Say 'Operation Aborted '
exit
end
if cpanel = 'ADB21S' then do
"TBSORT " ÅTABLE " FIELDS(TYPE,C,A,NAME,C,A)"
end
"VPUT (CPANEL) SHARED"
"SELECT CMD(%CMEDIT TABLE("ÅTABLE") SKELETON(JNADBDRA)"
"VERASE (CPANEL) SHARED"
Exit

zAnalytics
18
Sample CMEDIT
PROC 0 TABLE() TABLE02(NONAME) TABLE03(NONAME) TABLE04(NONAME) +
TABLE05(NONAME) SKELETON() MACRO() STCLIST()
IF &STCLIST ^= &STR() THEN DO
ISPEXEC SELECT CMD(%&STCLIST TABLE(&TABLE))
END
IF &MACRO ^= &STR() THEN DO
SET MACRO = MACRO(&MACRO.)
END
IF &TABLE ^= &STR() THEN DO
ISPEXEC TBQUERY &TABLE ROWNUM(ROWNUM)
END
ISPEXEC FTOPEN TEMP
ISPEXEC FTINCL &SKELETON
ISPEXEC FTCLOSE
ISPEXEC VGET (ZTEMPN)
ISPEXEC LMINIT DATAID(DATAID) DDNAME(&ZTEMPN)
ISPEXEC EDIT DATAID(&DATAID) &MACRO
ISPEXEC LMFREE DATAID(&DATAID)

zAnalytics
19
Sample Skeleton
&ADBJ1
&ADBJ2
&ADBJ3
&ADBJ4
&ADBJ5
)TB 66
)SET CMDSG = &DB2SYS
)IM CMDSNTEP
)DOT &TABLE SCAN
)SEL &CPANEL = ADB21S
DROP TABLESPACE &DBNAME..&NAME.;
)ENDSEL
)SEL &CPANEL = ADB21T
)SEL &TYPE NE A
DROP TABLE &CREATOR..&NAME.;
)ENDSEL
)SEL &TYPE = A
DROP ALIAS &CREATOR..&NAME.;
)ENDSEL
)ENDSEL
)SEL &CPANEL = ADB21VV
DROP VIEW &CREATOR..&NAME.;
)ENDSEL
)SEL &CPANEL = ADB21X
DROP INDEX &CREATOR..&NAME.;
)ENDSEL
COMMIT;
)ENDDOT
END

zAnalytics
20
The Catalog Tool Extentions
Commands directly in Panels (Maintenance)
– DIS Display in Table Panel
– ICS See Image Copy from table panel
– KCU Packages with Contoken/Last Used
External Commands
– DUT Homegrown Adhoc Utility Engine
– DUS Homegrown Utility Automation
– UHIS, RTSX Homegrown RTS Solution
– UNL, UNLO, LOA, XLOAD – Unload/Load on homegrown versions.
Other
– Suppression of Templates because we use SYSTEMPL as a good solution to keep
company defaults local
– DUT – Database Utility Tool

zAnalytics
21
JN --------------------------- DB2 Utility Generator --------------------------
Option ===>
Database ==> XDB05411 Userid - TEKSAS
Tablespace ==> % Time - 20:05
Partition ==> Terminal - 3278A
Part level ==> Y (Y/N) Pf keys - 24
Job-divide ==> N (Y/N) Subsystem- DBX0
Recall ==> Y (Y/N) Catalog - SYSIBM
Spanned ==> Y (Y/N)
Where ==>
Option Option
CHD Check Data REC - RECover
CHI Check Index RECV - RECover Value (Advanced)
IC Image Copy With Template REP - REPort
IIC Incremental Image Copy with template REO - Reorg
ICE Image Copy with Template i own name RES - Display REStrict/RESolve
INI INItial Load - Load Dummy RIX - Rebuild IndeX
LOA LOAd tables from UNL-function RUI - Recover Without indexes
LRSN Calculate LRSN RUN - Runstats
QUI QUIesce STA - Start database
RCA ReCAll tapes from VTV STO - Stop Database
REBIND Rebind all packages including DB/TS UNL - UNLoad tables
UNLO - UNload and Generate LOad
SE - SEtup Session-variables
Press END to end DB2 Utility generator.

zAnalytics
22
Useful ”hidden” Commands
CAT Catalog commands (Hyperjump)
– CAT K BATCH.PROGRAM1
– CAT T Creator.Table
PROMPT SEL ON – Shows the SQL
SQL – Jumps to the SQL Editor or executes a given sql-statement
LIKE ON/OFF (turn the wildvardsetting off/on)
SAVE ON/OFF (turn remember on/of)
B/BR Primary Command (uses BRIF interface to quickly show table content)
WSL – Jumps to Work Statement List
SSID <db2 subsys> change DB2 Subsystem
VER ON/OFF i Package panel

zAnalytics
23
Are We Happy ?

zAnalytics
24

Db2 update day 2015 managing db2 with ibm db2 tools svenn aage

More Related Content

What's hot (20)

Similar to Db2 update day 2015 managing db2 with ibm db2 tools svenn aage (20)

Recently uploaded (20)

Db2 update day 2015 managing db2 with ibm db2 tools svenn aage