IMS CT Test Envir - IMS UG November 2012 Phoenix

Data Management Tools

© 2012 IBM CorporationM Corporation

Cloning Test Systems –Cost, CPU, and Time Savings

Rosemary Shay

Email: [email protected]

Ron Bisceglia

Email: [email protected]

Data Management Tools – IMS Tools

© 2012 IBM Corporation2

Topics

� Why Do We Test

� Copying Data, the Fast Way

� Cloning Automation

– Cloning Terminology

– Challenges of Cloning

– Clone IMS systems

– Refresh IMS databases

� Additional Test Environments


© 2012 IBM Corporation

What Causes Outages

3

Based on extensive feedback from clients, we estimate that, on average, unplanned application downtime is caused: 20 percent of the time by hardware (e.g., server and network), OSs, environmental factors (e.g., heating, cooling and power failures) and disasters; 40 percent of the time by application failures including "bugs," performance issues or changes to applications that cause problems (including the application code itself or layered software on which the application is dependent); and 40 percent of the time by operator errors, including not performing a required operations task or performing a task incorrectly (e.g., changes made to infrastructure components that result in problems and incur unexpected downtime).

Thus, approximately 80 percent of unplanned downtime is caused by people and process issues, while the remainder is caused by technology failures and disasters. Improving availability requires a different strategy and set of investment choices for each of the three unplanned downtime categories.” -- Gartner Group



Database Downtime Drives Up Costs

Most organizations spend an extra $1.5M USD per year because of unplanned database downtime

Very few organizations have perfect or near-perfect datacenter uptime

�Only 3% of organizations have uptimes of 100%

� Only 4% of organizations have uptimes of 99.9%

Source: Aberdeen Group, 2010



Why is Test Data Management Important?� Improving business critical applications

– Improve application reliability

– Increase functionality

– Shorten time to market

– Reduce development and testing costs

� Sharing resources and skills for development and testing

– Setting up a test environment usually involves:

• Systems programmer

• DBA

• Application developer

• Tester

– Test databases are typically shared

� Meeting technical requirements

– Application testing done in a production replica

– Technical changes tested before production implementation

� Meeting regulatory requirements

– Protection of personal or sensitivity data

5



Copying Data, the Fast Way



Host Based Data Copy Options

� Volume copy options

–DFSMSdss (IBM)

–FDR (Innovation Data Processing)

–TDMF (IBM)

–FDRPAS (Innovation Data Processing)

� Data set copy options

–DFSMSdss (IBM)

–FDR (Innovation Data Processing)

Data copy processes use host based CPU and I/O facilitiesMore costly and slower than storage-based fast replication

Host-basedCopy Process



Storage Processor = Storage + Processor

�Storage – Non-volatile

– Computer components and recording media that

retain digital data

�Processor – Computing system that executes software

programs

�Storage Processors – Enable data movement through host-based API

– Allows CPU reduction from host-based systems

8

Host based APIs



What is Storage-based Fast Replication?

� For volume replication a relationship is established between a source and a target (Pair)

– Geometrically similar devices

� A near instant copy of a volume/data set at a specific point in time

– Builds a bitmap to describe the source volume

– After the bitmap has been created, the source and target

volume data can be used immediately

� Data movement (CPU and I/O) offloaded to storage processor

– Frees up resources on host processor

– No host CPU or I/O costs

� Consistency Groups

– Group of volumes copied at exactly the same point in time

while maintaining the order of dependent writes

–9

Storage Processor-based

Copy Process



Fast Replication Data Copy Options

�Volume based fast replication options– FlashCopy (IBM,EMC,HDS)

– SnapShot (IBM,STK)

– TimeFinder/Clone Volume Snap (EMC)

– TimeFinder/Snap (EMC)

– Mirror processes

•PPRC (IBM,EMC,HDS)

•TimeFinder/Mirror, SRDF (EMC)

•ShadowImage HUR (HDS)

�Data set based fast replication– Data set FlashCopy (IBM,EMC,HDS)

– Data set SnapShot (IBM,STK)

– TimeFinder/Clone data set Snap (EMC)

10

Fast copy processes offloaded to the storage processorNo host CPU or I/O resources

Fast ReplicationCommands from

z/OS



Database and Storage Integration

� Repurpose resources that you already own: Storage systems

– You already have a significant investment in storage systems

– Use storage systems to do more than you thought they could

• Simplify database administration tasks

• Speed up database administration tasks

• Reduce the costs of database administration tasks

� Database knowledgeable and storage aware

– Storage aware tools understand the language and the environment of the

database administrator

• Assists the DBA by doing research that the DBA would normally have to do.

• Translates database administration clone, copy and refresh functions into

commands used by the storage subsystem

• Provides simulation and undo options

• Provides a safe, user friendly interface between the DBA and powerful

storage utilities

– Allows the Storage Administrator to breath easier while the DBA uses storage

tools

11



Database and Storage Integration

MainframeDatabase Systems

Storage AwareDatabase Tools

Application and Database Management

Domain

Storage Administration and

Business ContinuityDomain

• Organizational Integration• New Backup Methods• New Recovery Strategies• Business Recovery Monitoring• Cloning Automation• Disaster Restart Solutions

SourceDatabase

Backup,Clone,

DR



Cloning Automation



Cloning Terminology

� A clone is an exact replica

– Clone an IMS system by volume

– Clone an IMS Database by data set

� IMS system cloning and database refresh

– The act of replicating the data, making the replica accessible, and then using

the replica in lieu of the original data

� IMS system cloning automation

– Clones a complete IMS system including all its databases

� IMS database refresh

– Refreshes specific IMS databases



IMS System ‘Cloning’ Today

Production LPAR

User Catalogs

IMS001

IMS System Data Sets

IMS002

Application Libraries

IMS003

RECON Data Sets

IMS004

Data Base Data Sets

IMS005

Test LPAR

User Catalogs

IMS001


IMS002


IMS003

RECON Data Sets

IMS004

Data Base Data Sets

IMS005



IMS Database ‘Cloning’ Today

Production IMS


RECON Data Sets

Data Base Data Sets


Test IMS


RECON Data Sets

Data Base Data Sets


Unload Load

Image Copy Recovery

Index Rebuild

???

z/OS



Challenges to Using Volume Fast Replication

� IMS system cloning is best done using volume replication

– Large amounts of data

– Volume replication much faster than data set replication

• No z/OS dataset allocation

� Volume data is replicated fast and easy but…

– How do you access the cloned data?

� Inherent Problems:

– VOLSERs may have the same volume names as the source

– Volume VTOC, VTOCIX and VVDS would reference the source VOLSERs

– Data set names are the same on source and target volumes

17



� Rapid Volume Reconditioning

– IMS Cloning Tool performs rapid volume reconditioning

and data set renaming on cloned system volumes to

solve the data access challenges

– Provides access to data sets on target volumes

• Renames the VTOC, VTOCIX, and VVDS to match the target volume

• Renames and catalogs all data sets to a new HLQ

– Solution to enable the use of the data sets on the target

volumes on the same image

• No more complicated cloning

• No more requirement for a separate image

IMS Cloning Tool - Provides Data AccessOn the Same or Shared LPAR



B.DSN1

B.DSN3

B.DSN2

VTOC

SourcePIMS01

Target TIMS01

VTOCIXVVDS

ICF User

Catalog

A1.CAT

A.DSN3

A.DSN2

A.DSN1

VTOCVTOCIX

VVDS

1. Target Volume is renamed.

2. Volume VTOC, VTOCIX and VVDS now reference the target VOLSERs.

3. Data sets are renamed and cataloged.

IMS Cloning Tool - Provides Data AccessOn the Same or Shared LPAR



Adjusts Target IMSTo Accommodate and Accept the Cloned Data

� RECONs data sets

– Data set names, IMS subsystem IDs, and VOLSERs are updated in the following RECON records: header record, database data set records,

online log records, and back-out records

� IMS PROCLIB and JOBS and user JCL libraries

– New values for IMSID, VOLSERS, and data set names in the JCL

members within these libraries

� MDA (MVS Dynamic Allocation) members for databases and system data sets, are updated to reflect the new data set names. Updates include:

– The RECON data sets

– Online Log data sets (OLDS)

– Write-ahead data sets (WADS)

� If IMS data sharing is involved, each additional IMS data sharing member is updated



IMS Cloning Tool Delivers Solutions� Leverages Storage-based fast replication

– Very fast -13TB in .4 seconds - no outage of source system

– Large systems cloned in less then 30 minutes

– Saves hours over traditional cloning

– Saves CPU and I/O processing

� Automation

– We condition the copied volumes for usability

– We condition the cloned IMS’s RECON, PROCLIB,JOBS, MDA

members – everything you need to bring up your cloned IMS

– Databases are copied, underlying data sets renamed, DBRC updated

– Repeatable process

With IMS Cloning Tool it is now easy and inexpensive to create cloned environments for testing, offloading

reporting, moving data or aiding in problem determinations.

21



Do I have to Use Storage-based Fast Replication?

� IMS Cloning Tool works with Host-based copy options

- Consider the availability requirements of the

source system

- Consider the amount of data being copied

� All the same test environments can be created

� Rapid volume reconditioning will be performed to make the copied volumes accessible

� The target IMS datasets will be conditioned

22

Host-basedCopy Process



Walking Through the Cloning Process



Pre-System Cloning Steps

� Review and Define Target System PROCLIB members

– Determine what components and configuration options you want

• IRLM, CSL, XRF, IMS Connect, etc.

� By default, because the clone is an exact replica of the source, the clone would have the same components as the source

– Review IMS.PROCLIB members (DFSPBxxx, etc.)

– IMS Configuration Manager can review and create new members



Production Database

IMSClone

IMS

SourceDatabase

Volumes

Clone

DatabaseVolumes

� Performs IMS system cloning automation

– Takes an existing IMS system (complete installation and system generation process completed) and creates a new, or cloned, IMS system from it without having to repeat the entire installation and system generation processes

– Simplifies IMS cloning processes

– Reduces cloning time and administration costs

� Leverages fast replication facilities to clone data

– Host-based copy can also be used

� Performs rapid volume reconditioning and data set renaming on cloned volumes to solve the data access challenges

– Target volumes retain their target volume label

– Renames the VTOC, VTOCIX, and VVDS to match the target volume

– Renames and catalogs all data sets to a new HLQ

� Adjusts target IMS to accommodate and accept the cloned data

– IMS RECONs, PROCLIB, JOBS, JCL, MDA members

IMS System Cloning Automation

IMS Cloning ToolCloning

Automation



IMS System Cloning Steps

Production IMS‘Source’

IMSClone

Target IMS

6

Start cloned IMS

1 IMS volumeselection

2

Volume copy3

A. Log SuspendB. Consistency

Group

4 Log Resume if 2A

7

Volume Recondition5

IMS

Source

Database

Volumes

Clone

Database

Volumes

Update cloned IMS meta data

8

Data set rename



Status of In-Flight Transactions

� An online cloning solution often results in transactions in flight

� These in-flight transactions, cloned to the target system result in the same target system action that would happen on the source system if it were to have died at that same time and then restarted

� The target restart is essentially an emergency restart of a failed system



IMSSystem Volumes

IMSTarget

IMS Clone

Volumes

IMSSource

IMS Cloning Tool Database Refresh

Automation

� Performs automated IMS database refresh operations

– Fast refresh of IMS databases

– IMS DB support (FF, HALDB, DEDB)

� Verifies source and target database compatibility

� IMS data copied using storage-based data set fast-replication

– Target takes up the same amount of space as the source

– Host-based copy can also be used

� Performs target system meta-data management

– Updates DBRC information for target databases

– Supports HALDB special requirements

• Primary and secondary indexes and ILDS can be copied to eliminate index and ILDS rebuild time

• Partition and reorg numbers will reflect the source RECON

Refresh IMS Databases Using IMS Cloning ToolDatabase Refresh Automation

Production Database



Refresh IMS Databases Using IMS Cloning Tool


IMSTarget

Target IMS

8

2 7

Create target IMS databasesif they do not exist

Data Set Copy Process

Start Target Databases

Update DBRC

for Target DatabasesDatabase Selection

Stop Source and Target IMS databases

Start Source IMS databases

3 Compatibility Check

4

5

6

1IMS



Additional Test Environments



IMS System Skeleton Clone

� IMS System Skeleton clone

– Clones an IMS system without cloning any data

– The replicated system is accessible and is usable in lieu

of the original system without requiring a system generation

– Contains all of the database and application definitions

– Creates an IMS system to test the apply process

31


© 2012 IBM Corporation32 32

Create an IMS Cloned System Skeleton

IMS

IMS System VolumesONLY

IMSClone

2 Perform IMS System Cloning Skeleton AutomationUsing System Volumes Only

3 Create Cloned IMS System Without Cloned Database Data


Target IMSTest new releases of IMS

IMS

SystemDataSets

IMS

SkeletonSystem

1 IMS System VolumeSelection

Data-baseDataSets

4

5

Install new IMS release or upgrades

Start Your IMS



What you can test with an IMS System Skeleton Clone

� IMS version migration

– Test your end to end apply process

• Test with full production RECON

• Re-clone for different environment configuration new/existing

• Can you bring your IMS up?

– Activate new functionality

• Migration to Striping and Buffers in 64-bit Storage

• RECON – changing MINVERS

• Dynamic Resource Definitions (DRD)

– Convert to DRD

– IMS Repository Test with full production RECON

� Manually clean up a RECON

– This may take more then one try; simply re-clone between tests

33


© 2012 IBM Corporation34 34

IMS

IMS System and Database Volumes

IMSClone

2 Perform IMS System Cloning AutomationUsing System DataSet Volumes Only

5

Refresh IMS Database A

3 Create Cloned IMS System Without Cloned Database Data

4

6 Test


IMS Database Refresh Environment

Refresh IMS Databases

Target IMS

IMS

SystemDataSets

IMS

SkeletonSystem

1 IMS System VolumeSelection Refresh IMS

Database B

A

B

A

B

IMS Cloned System SkeletonCopy Databases for Application Testing



IMS Cloned System SkeletonCopy Databases for Application Testing

� IMS version migration

– Focused Application and Database testing

• IVP validation steps

• IMS Exit Routines

• Systems with a history of having issues during past migrations

– Test new functionality

• FF- Dynamic Database Buffering

• DEDB – Secondary Index processing

• DEDB- Buffer Manager 64 bit Enhancements

� Determine if an issues is environmental or application related.

� Specific application full volume testing

35



Full System Clone

� Exact copy of production environment for the highest quality testing

• Test interfaces with other systems

• Application acceptance testing

• Load test high volumes

• Reset the environment by re-running the cloning process

� IMS version migration - performance

• zAAP/zIIP Accounting Log Records– Distinguish between CP (Central Processor) and zAAP/zIIP times

• Logging Enhancements– Striping – improved logging throughput

– Moved above 2 GB boundaries – ECSA saves

� Repeatable performance testing

• QCF (Queue Control Facility)

• IMS Performance Analyzer

36



IMS Gold Copy Cloned to Multiple Test Environments

37

IMS Source

Cloning Tool

Cloning Tool

IMSTarget A

IMSTarget B

IMSTarget C

IMS Target D

IMS Source

IMS Target

IMS Targets



Federated Gold Copy Cloned to Multiple Test Environments� Optim reduces and scrubs data from IMS, DB2, VSAM, and sequential files

� Cloning Tool creates target environments, refreshes databases and tablespaces

38

VSAM and sequential

IMS data

DB2 data

Cloning Tool

Optim

IMS/DB2Target A

IMS/DB2Target B

IMS/DB2Target C

IMS/DB2Target D

Reduceddata



IMS Cloning Tool V1.2 Preview

� ISPF user friendly interface

� IMS Database Refresh

– Data Masking

– Apply Logs after refresh process for databases consistency

– Support Database Quiesce

�System Cloning

– Enhance data set RENAME masking

– Allow storage groups to be used

– Support Preserve Mirror Flash Copy

– Support Incremental Flash Copy

– Support Space Efficient Flash Copy Target Volumes

– Support EMC Timefinder/Clone Volume Snap



IMS Cloning Tool Benefits

�Significantly reduced cost by providing quality test environments

– Creates many types of testing environments to meet your needs

– Creates exact copies of your production environments

– Easily leveraged in many test environment schemas

�Significantly lower CPU and time needed when cloning multiple systems and databases verses traditional methods

– Leverage Fast Replication

– Perform Rapid Volume Reconditioning

– Condition the IMS meta data

– It takes 30 minutes vs. 4 days on average (customer experience)

�Significantly reduced administrator time

– IBM IMS Cloning Tool does the administrative work for you

– Freeing up people and systems for other requirements

– It)



Questions

IMS CT Test Envir - IMS UG November 2012 Phoenix

Technology

Transcript of IMS CT Test Envir - IMS UG November 2012 Phoenix