Adabas Concepts of a Database for Mission-Critical Applications
Friedrich-Schiller-Universitt Jena Lehrstuhl fr Datenbanken und Informationssysteme Prof. Dr. Klaus Kspert February 1, 2011 Software AG Arno Zude, R&D ETS Adabas
Agenda
1) Software AG at a Glance 2) About the History of Adabas 3) Adabas Concepts
Data Model Data Structures Recovery
4) Accessing Adabas 5) New Developments 6) Success Factors
Feb 1, 2011 | University of Jena | 2
Software AG at a Glance About the History of Adabas Adabas Concepts ...
Software AG at a Glance
Founded in 1969 2nd largest German software company
Headquarters in Darmstadt
4th largest European software company Ca. 1.2 billion (109) revenue in 2010 Ca. 5,700 employees worldwide
>2,000 in Germany ca. 850 in R&D
Represented in 70 countries More than 10,000 enterprise customers Ca. 29% owned by Software AG Foundation
Feb 1, 2011 | University of Jena | 3
Software AG at a Glance About the History of Adabas Adabas Concepts ...
Software AG at a Glance Products
1st relational-like, transactional high-performance database One of the first 4th-generation programming languages
First Business-to-Business server and SOA integration platform First business process analysis & design platform
Feb 1, 2011 | University of Jena | 4
Software AG at a Glance About the History of Adabas Adabas Concepts ...
Software AG at a Glance A Global Company
R&D Labs in All Major IT Hotspots
Feb 1, 2011 | University of Jena | 5
About the History of Adabas Adabas Concepts Data Model ...
About the History of Adabas
Adaptable Database System Adabas development started in 1969
first on mainframe
Relational-like, transactional, multi-user Used across all industries
Banking, insurance, logistics, government, etc.
More than 2,000 customers worldwide Developers in Germany, US, UK
Darmstadt, Reston/VA, Denver/CO, Derby
Runs on mainframes and open systems
z/OS, z/VSE, z/VM, BS2000/OSD Unix, Linux (incl. z/Linux), Windows
Feb 1, 2011 | University of Jena | 6
Concepts Data Model Data Structures ...
Adabas Concepts Data Model
Adabas database
Multiple files in a database
Multiple records in a file
Multiple fields in a record
Database
File
Record
Linking different files via common fields
no pointers between records
like in relational model unlike in hierarchical, network models
Field
Adabas vs. relational terminology:
Adabas Relational Table Row Column Index Commit Rollback
Read & update operate on individual records
not sets of records
File Record Field Descriptor End Transaction Backout Transaction
Search operations create lists of record IDs
not protected against parallel update
Differences to relational model
Feb 1, 2011 | University of Jena | 7
Concepts Data Model Data Structures ...
Adabas Concepts Data Model II
Various data types for fields
Character, binary (Un)packed decimal numbers Fixed-point, floating-point Large objects (e.g., multi-media)
Fields can be defined to have multiple values
Multiple-value fields
Groups of fields can be defined to have multiple occurrences
Periodic groups
Related data is kept close together Efficient data access Differences to relational model
Feb 1, 2011 | University of Jena | 8
Concepts Data Model Data Structures ...
Adabas Concepts Data Model III
Fast read & search access via indexes
called descriptors
Several types of descriptors
(Simple) descriptor
One full field
Field = Jena Descriptor = Jena
Subdescriptor
Part of one field
Superdescriptor
Combination of parts of several fields
Field = 20110201 Descriptor = 2011
Hyperdescriptor
Combination of several fields Value(s) determined by user exit
Collation descriptor
Sort sequence defined by user exit
Field 1 = Kuespert Field 2 = 20110201 Descriptor = Kuespert__20110201
Flexible, efficient data access
Feb 1, 2011 | University of Jena | 9
Concepts Data Structures Recovery ...
Adabas Concepts Data Structures
Container datasets (disk files)
Fixed block sizes, direct access
Block = minimal I/O unit
ASSO
FCBs FDTs Index 1 IX 2
DATA
File 1 File 2
DATA Data storage
The actual payload data
ASSO Associator
File directory Field definition tables Indexes Free space management
Adabas Server
WORK Work file for temporary use
Journal of updates for restart recovery Search results
WORK PLOG
PLOG Protection log
Journal of updates for archive recovery Copied off for archival when full
Seq. PLOG
Feb 1, 2011 | University of Jena | 10
Concepts Data Structures Recovery ...
Adabas Concepts Data Structures II
Data Compression
Field data typically stored in compressed form
Character: No trailing blanks Numeric: No leading zeros Null-value suppression option
Compact storage, efficient access Data storage (DS)
Variable-length data records One or more records per block
Record moved to other block if it becomes too large to fit
Feb 1, 2011 | University of Jena | 11
Concepts Data Structures Recovery ...
Adabas Concepts Data Structures III
Address converter (AC)
Each record identified by its ISN
Internal Sequence Number A unique number, usually serial
AC maps ISNs to DS blocks
Index
Index maps descriptor values to ISNs Stored as inverted lists
B*-tree
Feb 1, 2011 | University of Jena | 12
Concepts Data Structures Recovery ...
Adabas Concepts Data Structures IV
Finding data records from index via address converter to data storage:
Feb 1, 2011 | University of Jena | 13
Concepts Recovery Accessing Adabas ...
Adabas Concepts Recovery
Restart recovery
Performed if database server fails
Power failure Canceled by operator System problem Software problem
When server is restarted
Protection data on WORK is used to
detect need for restart recovery repair inconsistent internal data structures (e.g., index trees) redo committed updates that were not written out prior to the failure undo uncommitted updates that were written to the database
Feb 1, 2011 | University of Jena | 14
Concepts Recovery Accessing Adabas ...
Adabas Concepts Recovery II
Archive recovery
Performed if database gets damaged
Disk failure Application misbehavior Software problem
To repair damaged database or file(s)
Restore archive copy of database/file(s) Use the protection data on the PLOGs to
replay all updates performed between the time the archive copy was created and the time the damage occurred
Alternative: Use the PLOGs to
roll back all updates between the current time and the time first damage occurred
Feb 1, 2011 | University of Jena | 15
Accessing Adabas New Developments ...
Accessing Adabas
Application programming interfaces to Adabas
Direct call interface
For example, from COBOL, C
Natural (Software AGs 4GL) Adabas SQL Gateway
ODBC JDBC
Adabas SOA Gateway
SOAP, REST
Feb 1, 2011 | University of Jena | 16
Accessing Adabas New Developments ...
Accessing Adabas Direct Call Interface
Set up Adabas Control Block and up to five Adabas buffers with
Adabas command to be executed Data to be searched, read, or inserted/updated/deleted
Example (COBOL):
MOVE MOVE MOVE MOVE MOVE MOVE MOVE 'S1' TO ADACB-COMMAND. 'I' TO ADACB-COP2. DBID-FNR TO ADACB-DBFILE. 'FIND' TO ADACB-CID. 'AA,8,A,AE,20,A,AJ1-3,20,A.' TO ADABAS-FORMAT-BUFFER. 'AE,9,A,D,AJ,5,A.' TO ADABAS-SEARCH-BUFFER. 'KUESPERT JENA ' TO ADABAS-VALUE-BUFFER.
CALL 'ADABAS' USING ADABAS-CB, ADABAS-FORMAT-BUFFER, ADABAS-RECORD-BUFFER, ADABAS-SEARCH-BUFFER, ADABAS-VALUE-BUFFER, ADABAS-ISN-BUFFER.
Feb 1, 2011 | University of Jena | 17
Accessing Adabas New Developments ...
Accessing Adabas Natural
is
Software AGs 4th-generation programming language for application development
Example for calling Adabas:
FIND PERSONNEL WITH NAME = ' KUESPERT' AND CITY = ' JENA' DISPLAY 'ID' PERSONNEL-ID 'Surname' NAME 'Location' CITY END-FIND
Feb 1, 2011 | University of Jena | 18
Accessing Adabas New Developments ...
Accessing Adabas Natural II
Feb 1, 2011 | University of Jena | 19
Accessing Adabas New Developments ...
Accessing Adabas SQL Gateway
Adabas SQL Gateway
Calling Adabas via an SQL interface
SELECT FROM WHERE AND personnel_id, name, city, PERSONNEL t1, PERSONNEL_ADDRESS_LINE t2 name = 'Kuespert' AND city = 'Jena' t1.isn_personnel = t2.isn_personnel
Executed by accessing only one Adabas file (table)
PERSONNEL_ADDRESS_LINE is periodic group in PERSONNEL file
Embedded into host programming language Calling Adabas via an ODBC or JDBC interface
Feb 1, 2011 | University of Jena | 20
Accessing Adabas New Developments ...
Accessing Adabas SOA Gateway
Adabas SOA Gateway
Calling Adabas via a request to a Web service URL
http://ibm2.software-ag.de:7014/Adabas_220_88?WSDL http://ibm2.software-ag.de:7014/Adabas_220_88?GET&A1=adabas2006
Calling Adabas via a SOAP request Callable from programming languages, Web browser, Excel, Word, etc.
Feb 1, 2011 | University of Jena | 21
New Developments Success Factors
New Developments
Application Java .NET 3GL Natural
Cross-platform
Administration Security
External e.g. RACF
SQL
Java
XML
SOA
Data Access
Tamino XML Server
Text Retrieval Cluster
Availability Load Balancing
Monitoring Statistics Utilities
Performance
Windows, Unix. Linux, Mainframe
Partitioning Save/Restore
Data Organization
Data Management
Replication Transaction
Data Distribution
VSAM
DL/1
DB2
Apps
Feb 1, 2011 | University of Jena | 22
New Developments Success Factors
New Developments Adabas Cluster Services (2002)
z/OS A
User
XE S
CF
Lock
XES
z/OS B
User
Cache
XCF
Adabas Server
Adabas Server
DB
WORK PLOG Seq. PLOG
DB
PLOG WORK
Feb 1, 2011 | University of Jena | 23
New Developments Success Factors
New Developments Event Replicator for Adabas (2005)
Replication of production data Close to real time From Adabas (source) To
Adabas (destination) relational databases data warehouses applications
Integration Platform
Data Transformation, Enrichment and Routing
RDBMS
Database-to-Database Replication
JMS Messaging
Data Warehouse Business Intelligence
and Reporting
Applications Multi-Target and Application Replication
Feb 1, 2011 | University of Jena | 24
New Developments Success Factors
New Developments Data Archiving for Adabas (2010)
Archival of production data Resource savings
Automation Lower-cost hardware
X Production Database Archive Vaults Business Applications
Regulatory compliance
Storage Robust search capabilities
Extraction Filter
Performance
Relieve pressure on production database
Recall
Archive Database
Archive Vaults/Files
Feb 1, 2011 | University of Jena | 25
New Developments Success Factors
New Developments Data Masking for Adabas (2011)
Generation of Test data
production-like data for
application testing end user training demonstration research Sensitive Data
Automated process
Less resources to create test data
Test Data Data Masking
Regulatory compliance
Privacy Security
Application quality
Production versus generic data
Data Masking
Feb 1, 2011 | University of Jena | 26
Success Factors
Success Factors
Flexible data model
Supports relational structures, entity-relationship model, others
Excellent performance Key factor
Helped by data model, record-level locking, exploitation of system functions Economical use of disk space via data compression, saving on I/O operations
High availability
Little downtime for maintenance (planned outages) Quick & reliable recovery (unplanned outages)
Simple administration
A few database administrators take care of many Adabas databases
Several platforms
Present across mainframes and open systems
Feb 1, 2011 | University of Jena | 27
Adabas
A database for mission-critical applications
Feb 1, 2011 | University of Jena | 28
