IBM's DB2 is a "Relational Database Management System" (RDBMS), where the "database" contains the data you want to work with, the "management" is what you want to do to the data, and the "system" means that DB2 is a collection of computer software programs that perform this particular type of task. DB2 is designed to make the storage and analysis of data easier.

DB2 can be used to manage a wide variety of data, such as the results of research experiments, records of purchases for agency administrators, or the locations and serial numbers of pieces of equipment. DB2 is especially effective at handling data that need to be stored securely, data that need to be manipulated and analyzed interactively, and data that are the basis for one-time and recurring reports. Data input, data update, data storage, data security, data extraction and analysis, and report generation are all facilities of DB2.

DB2 runs on an MVS server at the Computer Center. The DB2 server is a high-performance database engine capable of handling large amounts of data and many concurrent users. It has comprehensive backup and recovery facilities and excellent security. DB2 houses a variety of NIH enterprise-wide data including administrative, grants, personnel, and research data.

Any registered user of the Computer Center can create, access, and manipulate data in DB2. DB2 data is accessible interactively via client/server connectivity (including the Web) or an online (full screen 3270) connection. It can also be accessed by batch jobs.

A relational database management system is one in which the data can be referenced in terms of its content, without regard to the way the data are actually stored. The database is perceived by users to be a collection of tables. Each piece of data can be referenced independently, without worrying about where it physically exists. Indexes provide efficient access to data in large tables. Multiple records can be inserted, updated and deleted at the same time, and many users can access and update data concurrently.

In DB2, data is stored in the form of tables (relations) comprised of rows (records) and columns (fields). All access to and manipulation of data in DB2 is accomplished via Structured Query Language (SQL). SQL provides a relatively consistent and simple English-like language for retrieving (selecting) and manipulating (inserting, updating, and deleting) data. Furthermore, SQL provides facilities for defining tables and controlling access to them. Users can create formatted reports containing information selected in many ways from the results of their queries.

The Computer Center offers DB2 to its users in an "open shop" environment where users create and control their own database objects. DB2 is fully compatible with the complex hardware and software environment of the Computer Center.

At the Computer Center, DB2 subsystems operate under their own copy of TSO. Users can enter TSO commands and use ISPF within their DB2 sessions. This makes it convenient to examine job output, check on data set names, edit data sets or determine the status of a job during an online DB2 session.

Online and batch application programming facilities allow programmers to embed SQL statements in C, PL/I, COBOL, FORTRAN, and Assembler programs, and have programs in those languages build and execute SQL statements. Client/server access allows an almost unlimited application development environment, including the C/C++ language, 4GL environments and the web. This capability allows users to create custom-tailored interfaces to DB2 for their applications and to perform sophisticated data validation as it is entered.

Products and Versions

Vendor	Product	Version
IBM	DB2 Development (DSND)	V5
IBM	DB2 Production (DSNP)	V5
IBM	DB2 NIH Data Warehouse (DSNW)	V5
IBM	QMF	V3.3
IBM	DB2PM	V5
Neon Systems	Shadow Direct ODBC	V4.01.00
Neon Systems	Shadow Web Server	V4.01.00
BMC	LoadPlus	V4.2.00
BMC	Catalog Manager	V5.4.02
BMC	Change Manager	V5.4.02D
Oracle	Transparent Gateway to DB2	V4.1.0

Benefits of having a development-level (DSND) and production-level (DSNP, DSNW) environments

• The creator (i.e., high-level qualifier) in SQL does not have to change between the DSND and DSNP/W subsystems.
• Provides easy migration between the DSND and DSNP/W subsystems.
• The same JCL can be used with minor changes.

Differences between the development-level (DSND) and production-level (DSNP, DSNW) environments

• Dataset Migration
  DSND datasets migrate, DSNP/W datasets do not migrate.
• Space Allocation DSND - maximum primary (PRIQTY) for a tablespace or indexspace is 850 cylinders (approx. 600000 KB). DSNP/W - maximum allowed by DB2 (2GB)
• The same JCL can be used with minor changes.