The Evolution of DB2 Application Development

White Paper

The Evolution of DB2 Application Development

Written by, Jim Wankowski

Quest Software, Inc.

© Copyright Quest® Software, Inc. 2007. All rights reserved.

This guide contains proprietary information, which is protected by copyright. The software described in this guide is furnished under a software license or nondisclosure agreement. This software may be used or copied only in accordance with the terms of the applicable agreement. No part of this guide may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording for any purpose other than the purchaser's personal use without the written permission of Quest Software, Inc.

WARRANTY

The information contained in this document is subject to change without notice. Quest Software makes no warranty of any kind with respect to this information. QUEST SOFTWARE SPECIFICALLY DISCLAIMS THE IMPLIED WARRANTY OF THE MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. Quest Software shall not be liable for any direct, indirect, incidental, consequential, or other damage alleged in connection with the furnishing or use of this information.

TRADEMARKS

All trademarks and registered trademarks used in this guide are property of their respective owners.

World Headquarters 5 Polaris Way Aliso Viejo, CA 92656 www.quest.com e-mail: [email protected] U.S. and Canada: 949.754.8000

Please refer to our Web site for regional and international office information.

Updated—April, 2007

CONTENTS

INTRODUCTION................................................................................. 2 DB2 ................................................................................................... 3 A NEW LANGUAGE IS BORN............................................................... 4

UNIVERSAL DATABASE............................................................................4 CLIENT/SERVER ................................................................................ 5 THE CHANGING IT ENTERPRISE ........................................................ 6 THE MODERN DAY DEVELOPER .......................................................... 7 TOAD® FOR DB2................................................................................ 8 REFERENCING OBJECTS: THE DEVELOPER’S PERSPECTIVE................ 9

DATABASE BROWSER .............................................................................9 ACCESSING DATA: IT’S MORE THAN JUST WRITING A SELECT STATEMENT...........10 DATA EDITOR ....................................................................................11 EXPORT WIZARD ................................................................................12 MASTER DETAIL BROWSER.....................................................................12

BUILDING AND TESTING SQL CALLS FOR THE MAINFRAME............. 13 SQL MODELER...................................................................................14 QUERY EDITOR ..................................................................................14 BUILDING SQL ..................................................................................15 OPTIMIZING SQL................................................................................16

CONCLUSION................................................................................... 18 ABOUT THE AUTHOR........................................................................ 19 ABOUT QUEST SOFTWARE, INC. ...................................................... 20

CONTACTING QUEST SOFTWARE ..............................................................20 CONTACTING QUEST SUPPORT ................................................................20

1

INTRODUCTION

In the 1970s, the majority of data was stored in flat file formats such as VSAM or in a hierarchical database management system (DBMS) called information management system (IMS). These applications were running exclusively on large mainframes. Another format for databases, the Network model, was similar in concept to the relational model, but never gained much acceptance. IMS was known for its fast transaction speeds but was very tedious to program and to administer; its limited hierarchical model could not handle more than one relationship. During this time, IBM was also developing a new type of database based on Edgar Codd’s relational model developed in the late 60s. This database used a flexible programming language called Structured Query Language (SQL). This whitepaper will discuss how DB2 application development has evolved over the last three decades and the lessons developers can learn from technology innovations.

2

DB2

In 1983, IBM introduced its next generation relational database --Database 2, or better known as DB2. Although DB2 was available in the fall of 1984, IT organizations were slow to adopt this innovative technology due to its relatively high overhead, and its slow transaction speed compared to IMS. The MVS (multiple virtual storage) boxes of the day just didn’t have the horsepower to run large volumes of data through DB2. As mainframe hardware matured through the 1980s, DB2 grew to be the database of choice for many companies.

The 1990s saw an explosion of growth for DB2 as the databases’ speed, reliability, and functionality was vastly improved. This coincided with a dramatic increase in mainframe horsepower. Currently, IT organizations are moving to the existing IBM flagship database and the standard RDMS for mainframe applications.

Figure 1: Mainframe DB2 Timeline

3

A NEW LANGUAGE IS BORN

The majority of programming in the 1970s and 1980s was done in COBOL, PL/I, Assembler, or FORTRAN. Developers and end users used 3270 “green screen” dumb terminals. Program calls accessing data either in flat files, VSAM, or in IMS were limited in scope. Applications of the day were not very flexible and ad-hoc queries were inconceivable. Predicate logic or the ability to handle multiple relationships did not exist. Data had to be retrieved and then collated into the appropriate result set programmatically. With the release of DB2 and SQL, application developers acquired new programming concepts. A program call with multiple pages of code was condensed to a single SQL call. The flexibility of the new programming language simplified programming, but it also created new performance issues. The language could generate a complex query/result set all from a single call, but if the SQL statements were not coded efficiently, horrendous performance would result. The other advantage to the SQL language is that the database calls could be created and tested independently from the applications. The SQL statement could be written, executed, and analyzed without someone having to compile and execute an application program.

Universal Database

In 1989, IBM released a version of DB2 designed to initially run on its OS/2 operating system. The new version experienced name changes such as DB2/2 and DB2 Common Server (DB2/CS), but in an effort to capitalize on the momentum of DB2 running on the mainframe, a new name arose; the universal database (UDB) to refer to DB2 across all platforms. IBM also began synchronizing the release numbers starting with V6. DB2 never gained much market share in the ‘90s when it ran on distributed platforms. This changed around V7.0 when IBM began to heavily market DB2 to compete with Oracle. Success and growth occurred with the adoption of DB2 on Linux, Unix and Windows.

Figure 2: DB2 for Linux, Unix, and Windows

4

CLIENT/SERVER

A completely new concept emerged in the early ‘90s when DB2 faced explosive growth. Windows-based workstations communicating with UNIX servers via networks were believed to be the new generation of data processing that would effectively eliminate the need for mainframes in three to five years! This new technology would promise the same results as the big clumsy, expensive mainframes for a fraction of the cost. The novel concept gave birth to other relational databases designed specifically for these types of platforms. Oracle, teetering on the verge of bankruptcy in the late ‘80s, saw a huge boost in sales, along with Sybase and SQL Server. Huge learning curves arose for these new environments, considering the lion’s share of IT professionals were primarily mainframe-oriented. A mass exodus of mainframe expertise occurred as technologists hoped to move over to this new technology. The advanced concept drove companies to spend exorbitant amounts of money on client server systems only to have them fail due to limitations of the workstations, networks and servers.

5

THE CHANGING IT ENTERPRISE

Toward the end of the ‘90s and into the present, the typical corporate enterprise network began to diversify into a much more distributed environment. As Unix and Windows servers grew more powerful and reliable, mission-critical data started to get pushed to multiple DBMSs running on various platforms. This, along with the Internet, required newly written applications since they completely changed the way companies did business. Web-based or other distributed types of applications such as Websphere are now the order of the day. The prediction of an obsolete mainframe proved false in the current IT world, and instead the mainframe has become an integral data server for very large, complex enterprises. It is estimated that over 75 percent of the world’s business data resides on mainframes. The end-user e-business applications tend to be Unix-based, but the mainframe remains the core data repository. The Unix server still cannot match the mainframe’s capacity, reliability, security, and processing power.

6

THE MODERN DAY DEVELOPER

Developers today typically code in object-oriented languages such as C, C++, JAVA, or .NET with graphical user interface (GUI) front ends. These new applications may extract data from multiple sources including the mainframe via dynamic SQL calls. The current generation of developers has little or no concept of mainframes or 3270 interfaces -- instead, they are coding Windows, or internet-based applications accessing mainframe data. These developers need GUI-based tools that provide them with the ability to reference objects, access data, and build or test SQL calls off the mainframe.

7

TOAD® FOR DB2

Toad® for DB2 provides a proven, intuitive tool that improves user productivity and code quality. Users of all skill levels now have a single console for managing objects, creating and debugging stored procedures, generating optimized SQL, and benchmarking performance for DB2 running on LUW and z/OS. Skills migration is effortless for developers and DBAs who are familiar with Toad on other database platforms.

Figure 1: Connection manager allows simple log-on to DB2 z/OS and LUW environments

8

REFERENCING OBJECTS: THE DEVELOPER’S

PERSPECTIVE

When developers develop a new DB2 application, they need to reference objects such as tables, views, indexes, synonyms, and aliases from the DB2 catalogue. In the early days of DB2, developers typically used the SQL processing using file input (SPUFI) query facility. This 3270 based-tool enables users to code and execute queries dynamically. Referencing objects requires a fairly high level of mainframe navigation knowledge -- well beyond the scope of the typical Windows developer. Nowadays, the ideal scenario would offer a Windows interface, which allows developers to view the listed objects, and build their SQL calls directly in context from the objects. Instead of requiring the manual coding of the SQL calls, this eliminates the time-consuming need for the DBA to run reports listing the objects.

Database Browser

Toad’s database browser provides an intuitive interface for accessing all information stored in the DB2 catalog. Developers have easy access to column structures, indexes, views, referential integrity, and other features. Users can apply filters to all object levels to display only relevant information.

Figure 2: Database browser provides easy access to all relevant catalog information

9

Accessing Data: It’s More Than Just Writing a SELECT Statement

Developers require a realistic sampling of data to adequately test their applications as they develop new ones. The native facilities available in DB2 are limited to JCL batch jobs which run IBM’s DSNTIAUL program or the UNLOAD utility to dispatch data. The UNLOAD utility would need to facilitate a batch LOAD job to import the data into test tables. This goes beyond the abilities of the typical GUI developer. Once again, it may require the DBA’s involvement to run these jobs. A simple point/click method to export/import data is ideal. Another good feature for developers is the data editor. This allows the developer to manipulate the test data for specific test cases. These types of tools are commonly available for Unix-based DBMSs such as Oracle and SQL Server, but are hard to find for the z/OS platform.

Toad for DB2 provides robust functionality for manipulating test data:

Figure 3: Simple to use data-editing interface

10

Fig 4: Simply drag a column into the blue window to group data

Data Editor

The data editor function allows you to easily view and manipulate data by simply keying over a data value. It also fully supports LOB data. The drag-and-drop GROUPBY feature makes it easy to organize data for simpler viewing.

11

Export Wizard

The Export wizard extracts data and places it in to multiple formats, such as comma delimited, XML, HTML, or Excel.

Fig 5: Export wizard allows exporting of data to multiple formats

Master Detail Browser

The master detail browser displays a parent row and all its related child rows in a referentially tied table.

Fig 6: Master detail browser

12

BUILDING AND TESTING SQL CALLS FOR THE

MAINFRAME

A solid understanding of SQL fundamentals is the most critical factor in the application development lifecycle. A SQL statement can be coded in hundreds of ways and achieve the same result set with drastically different performance. Many developers with rudimentary knowledge of SQL need help with syntax and proper coding rules of thumb. The proper education is essential for developers in the explanation process to interrogate their queries and ensure effective coding. If overlooked in the application lifecycle, poorly written SQL calls will require DBA intervention when programs go into production. Once the SQL calls are built and tuned, developers should test them in a realistic environment to avoid performance issues that may arise when applications move to production. Many novice mainframe developers have a “desktop” mentality and are used to dealing with small Unix- or Windows-based databases. The common mistake is to test an application with limited data or an unrealistic user load. This leads to programs crashing when exposed to the typical transaction and data volumes of a production mainframe.

Toad for DB2 provides deep SQL development functionality, including automated SQL generation, syntactical help, customizable SQL templates, and integration with SQL optimization, and scalability testing tools.

13

SQL Modeler

The SQL Modeler option allows you to easily build SQL statements directly from a selected table. Window drop-down menus display aggregate functions and sort options for completing the SQL structure. You can select multiple tables to generate complex constructions. Generated queries can be reviewed, executed and explained.

Figure 7: Building a join statement with SQL Modeler

Query Editor

The Query Editor provides a robust facility for generating, formatting, and testing a query. You can free-form write a query or import it in from a file. Several shortcut functions help you to generate new queries, such as code snippets (syntax generators) and customizable templates for automatically producing specific lines of code. The tool can also generate syntax to embed SQL into applications written in C++, Delphi, Pearl, or JAVA code.

Figure 8: Functionality available in the Query Editor

14

Building SQL

With the Query Editor window, you can free-form write SQL, import from a file, and use the available shortcut commands such as code snippets and auto-replace templates.

Figure 9: Using the query builder option to generate SQL

Figure 10: Code snippets provide shortcuts to generating syntax

15

Optimizing SQL

There are two options for optimizing DB2 z/OS SQL within Toad.

Standard Explain – Built into Toad, the standard explain provides an animated visual display of the access path in multiple formats.

Fig 11: Built-in animated explain display

16

Advanced Explain – The advanced explain utilizes the z/OS SQL tuning module from Quest Central for DB2. This module provides a detailed overview of the access path, catalog statistics, and interactive tuning with the SQL compare feature.

Figure 12: Advanced explain

Figure 12: “Scenario” compare for interactive tuning

17

CONCLUSION

Once thought to face extinction, the mainframe now provides the backbone of data processing power for very complex distributed environments. The need to rewrite legacy green screen applications or to create completely new ones to address the modern day business need results in a shortage of mainframe programming expertise. Developers with disciplines other than the DB2 mainframe need a complete set of productivity tools for accessing mainframe objects, building efficient SQL, and performing pre-production stress testing.

18

ABOUT THE AUTHOR

Jim Wankowski is the DB2 Technology Specialist at Quest Software. He has over 20 years of development and DBA experience with DB2. Jim participated in the original beta program for DB2 in 1984 and is well known in the DB2 community. He has written articles for DB2 Magazine, and regularly presents at IDUG conferences, regional DB2 user groups, and vendor seminars worldwide.

19

ABOUT QUEST SOFTWARE, INC.

Quest Software, Inc. delivers innovative products that help organizations get more performance and productivity from their applications, databases and Windows infrastructure. Through a deep expertise in IT operations and a continued focus on what works best, Quest helps more than 18,000 customers worldwide meet higher expectations for enterprise IT. Quest’s Windows Management solutions simplify, automate and secure Active Directory, Exchange and Windows, as well as integrate Unix and Linux into the managed environment. Quest Software can be found in offices around the globe and at www.quest.com.

Contacting Quest Software

Phone: 949.754.8000 (United States and Canada) Email: [email protected]: Quest Software, Inc. World Headquarters 5 Polaris Way Aliso Viejo, CA 92656 USA Web site www.quest.com

Please refer to our Web site for regional and international office information.

Contacting Quest Support

Quest Support is available to customers who have a trial version of a Quest product or who have purchased a commercial version and have a valid maintenance contract. Quest Support provides around the clock coverage with SupportLink, our web self-service. Visit SupportLink at http://support.quest.com

From SupportLink, you can do the following:

Quickly find thousands of solutions (Knowledgebase articles/documents).

• Download patches and upgrades.

• Seek help from a Support engineer.

• Log and update your case, and check its status.

View the Global Support Guide for a detailed explanation of support programs, online services, contact information, and policy and procedures. The guide is available at: http://support.quest.com/pdfs/Global Support Guide.pdf

20