Transcript Change_Management

Managing Database Change
with Data Modeling
Bert Scalzo, PhD
[email protected]
© 2008 Quest Software, Inc. ALL RIGHTS RESERVED.
Bert Scalzo …
Database Expert & Product Architect for Quest Software
Oracle Background:
•
Worked with Oracle databases for over two decades (starting with version 4)
•
Work history includes time at both “Oracle Education” and “Oracle Consulting”
Academic Background:
•
Several Oracle Masters certifications
•
BS, MS and PhD in Computer Science
•
MBA (general business)
•
Several insurance industry designations
Key Interests:
•
Data Modeling
•
Database Benchmarking
•
Database Tuning & Optimization
•
"Star Schema" Data Warehouses
•
Oracle on Linux – and specifically: RAC on Linux
Articles for:
•
Oracle’s Technology Network (OTN)
•
Oracle Magazine,
•
Oracle Informant
•
PC Week (eWeek)
Articles for:
•
Dell Power Solutions
Magazine
•
The Linux Journal
•
www.linux.com
•
www.orafaq.com
2
Books by Bert …
Coming in 2009 …
Out Now …
2nd Edition
Coming Soon
3
Agenda
• Define Database Change Management
• Examine Common DBA Options for Managing
Database Structural Changes
• Examine the World of Modeling
• Why Data Modeling Often Makes Sense as the
Change Management Methodology
• Examples using Toad Data Modeler
Presentation Objectives
To demonstrate:
• Why a change management system is essential
• That database structural changes can be
accomplished in many ways
• How data modeling tools can be an essential part of
database change management
• The advantages of data modeling in change
management
What is Database Change Management?
…Ensuring that the source and target databases
are in sync after changes to the environment:
•
•
•
•
•
Development database
Test database(s)
Production database
Distributed databases
Replicated databases
Task is harder than you may think!
Common Database Change Management Tasks
•Create/Drop tables
•Add/Drop column constraints
•Add/Drop table columns
•Modify column constraint logic
•Resize table columns
•Add/Drop table & index partitions
•Add/Drop index columns
•Add/Drop object grants and roles
•Reorder index columns
•Resize or Relocate database objects
Change Management Can Be Tedious…
• Some database sync tasks require lots extra,
indirect work (depending on the Oracle server
version)
– Drop table columns
– Resize (i.e. shorten) table columns
– And a few others …
• Therefore, the DBA must
– Be aware of all these scenarios
– Write extended alteration scripts
– E.G. copy table, copy data, rename, grants
Common Change Management Options
• Direct to Database
• One Way DDL Scripts
• One Way DDL Scripts + Version Control
• Two Way DDL Scripts
• Two Way DDL Scripts + Version Control
• DB Schema Managers
#1: Direct to Database
OEM
TOAD
DBA
Database
SQL
Plus
•Pros:
•Method of least resistance (i.e. easy)
•Manually alter the database directly
•Often use highly visual tools like TOAD DBA and OEM
•Requires no extra tools (no extra $ or learn curve)
•Cons:
•The database is the Master
•Totally ad-hoc database mods can be problematic
•Extended alters require lots manual DBA time
•History of structural changes is lost
#2: One Way DDL Scripts
DDL
Script
Editor
SQL
Plus
•Pros:
•An Old Timer’s favorite
•Manually edit a Master DDL script
•Requires no extra tools (no extra $ or learn curve)
•Cons:
•The DDL Script is the Master
•Can forget & thus loose ad-hoc database mods
•Extended alters require lots manual DBA time
•History of structural changes is lost
Database
#3: One Way DDL Scripts + Version Control
DDL
Script
Editor
Source
Code
Control
Check-In
SQL
Plus
Check-Out
•Pros:
•Manually edit a version of Master DDL script
•Requires source control tool (free on UNIX)
•Addresses history shortcoming of prior method
•Cons:
•The DDL Script is the Master
•Can forget & thus loose ad-hoc database mods
•Extended alters require lots manual DBA time
Database
#4: Two Way DDL Scripts
Orig
DDL
Script
File
Diff
Tool
Editor
Gen
DDL
Script
DDL
Extract
Tool
New
DDL
Script
SQL
Plus
•Pros:
•Manually create DDL script based upon compare
•Requires DDL extract tool (can be SQL script)
•Cons:
•The DDL Script is the Master
•Can miss & over-ride ad-hoc database mods
•Extended alters require lots manual DBA time
•History of structural changes is lost
Database
#5: Two Way DDL Scripts + Version Control
Check-Out
Source
Code
Control
DDL
Script
Check-In
File
Diff
Tool
Editor
Gen
DDL
Script
DDL
Extract
Tool
New
DDL
Script
SQL
Plus
•Pros:
•Manually create DDL script based upon compare
•Requires DDL extract tool (can be scripts)
•Requires diff & source control tools (free on UNIX)
•Addresses history shortcoming of prior method
•Cons:
•The DDL Script is the Master
•Can miss & override ad-hoc database mods
•Extended alters require lots manual DBA time
Database
#6: DB Schema Managers
Quest
Schema
Manager
TOAD
DBA
DB
Def
File
Compare
and
Sync
Database
Deltas in
Database
Compare
and
Sync
Database
•Pros:
•Automatically create DDL script or directly apply selected changes
(i.e. sync) based upon compare
•Requires DB comp & sync tool, but then no other tools
•Addresses history shortcoming of prior methods
•Extended alters as easy as any other modification
•Cons:
•Disconnect between data models and database
Data Modeling as a Change Management
Methodology
• Works off the meta-data documenting the business
requirements (i.e. the true source code so to speak)
• Requires using just one tool – the data modeling tool
• Eliminates having multiple steps and branch points
• Sometimes picture of what’s changed is more readily
apparent and understandable than the alternatives
• By very nature, keeps database & its requirements in
sync!
The World of Modeling …
• End-user
• IT Partner/Liaison Business Process Modeling
(BPM)
• Business Analyst
• Improve process efficiency
• Define/document Bus. Processes
- create correct and complete
application requirements
• System Architect
• System Analyst
• App Developer
• Support for all UML diagrams
- Analyze requirements
- Design application
• Reverse/forward engineer code
• Bus. Analyst
• Data Architect
• Data Analyst
• DBA
• DB Developer
• DB Architect
Object-Oriented Modeling
(OOM - UML)
Conceptual Data Modeling
(CDM – E/R)
Physical Data Modeling
(PDM)
• Identify all data & relationships
- E/R (Entity/Rel’ship) diagrams
- DB independent view
• Business Rules?
•
•
•
•
DB-specific model
Reverse engineer existing DB
Create/Update DB from model
Data Warehouse Modeling
#6: Data Modeling Tools
Toad
Data
Model
Toad
QDesigner
Data
QDesigner
Data
Modeler
Modeler
Compare
and
Sync
Database
Deltas in
Database
Compare
and
Sync
Database
•Pros:
•Automatically create DDL script or directly apply selected changes (i.e.
sync) based upon compare
•Requires DB compare & sync tool, but then no other tools
•Addresses history shortcoming of prior methods
•Extended alters as easy as any other modification
•No disconnect between data models and database
Handle Like Programming Code
Check-Out
Developer
IDE
Source
Code
Control
Check-In
Check-Out
Modeling
IDE
Model
Repository
Check-In
Compiler
Object
Code
Linker
Compare
Executable
DDL
Script
Sync
Database
Compare & Sync Combinations
• Can compare & sync
–
–
–
–
–
–
–
–
Logical to Logical
Logical to Physical
Physical to Logical
Physical to Physical
Script to Physical
Database to Physical
Database to Database
Version to Version
MOVIES Data Model
Here we go …
• “Best laid plans of mice and men” 
• Real life is not as rigid or structured as we plan
• Thus even when adopting a data modeling change
management approach, two way compare/sync will
still be critical (need to catch exceptions to method)
• In following demo:
– Will change some live database objects using
Toad (e.g. alter table, create table, drop table)
– Will change some source code or data modeling
items (e.g. alter entity, create entity, drop entity)
– See how modeling tool can find and fix these!
Parting Thoughts
• Data modeling can be an invaluable method for
managing and addressing change in your DB
environments
• There are many other issues data modeling can
address:
– Modeling accurate business requirements
– Transforming logical into physical models
– Breaking data models into sub-models
– Round-trip Engineering:
• Conceptual -> Physical Model compare and sync
• Physical Model -> Database compare and sync
– Repository-based collaborative modeling
– Horizontal and Vertical Partitioning
– Data Warehousing (Star Schema design)
– Object-Relational Mapping
Questions and Answers …
Thank You
Presenter:
Bert Scalzo: [email protected]