This blog illustrates the history and development of the IT from the first ideas and machines in the 19th century to modern times.

Nowadays we talk about IT. Information Technology is the right term to describe the abundance of hardware and software.

The author of this article has been active in the field of data processing since 1968. Yes, back then it was still called data processing, or rather electronic data processing (EDP). Programmers or computer specialists from the German city Essen could be easily recognized by their license plate: E-DV nnn (EDV is the German word for EDP).

The first model of an analytical computing machine was introduced by Charles Babbage in 1837. It took roughly a hundred years until Konrad Zuse built the Z1 in 1941 and the Z4 in 1945. Those were the first programmable universal computer models.

That was the beginning of the age of data processing!

Sequential Processing

In the 60s and 70s data was captured and processed with punched tapes and cards. The data stored on electronic media (magnetic tapes, hard disks, and drum storage) were sequential files. The data was processed sequentially, record after record. Records could be of fixed or variable length. Multiple records were collected in one block. Certain records could not be accessed directly. The record had to be found by searching the sequential file. Some companies still have data relics from that time in productive use.

The sequential files were not suitable for efficient and quick processing of mass data. In the early 70s more efficient and precise data management systems were already in development.

Direct Addressing

Databases created in that time were based on Direct Access Methods (DAM): UDB (Universal Database) and DBOMP (Database Organization and Maintenance Processor) which were mainly used for managing bills of materials. Both systems were designed to use direct access storage devices (DASD) with a limited number of cylinders.

These databases were developed by IBM. The development was discontinued with the launch of new and bigger disk systems.

In the 1960s IBM released a database called ICS (Information Control System) that was renamed into IMS (Information Management System) in 1969.

Let‘s have a closer look at the popular mainframe database and file systems:

IMS and DLI

IMS is used by countless big companies as database system for applications that require high transaction rates and a high system availability.

IMS (also known as DLI for the operating system z/VSE) is a hierarchical database. Records are stored as a tree structure. Each record has a parent, while the root segment at the top does not have a parent.

The underlying data structure of IMS and DLI is VSAM.

ADABAS

Adabas was developed by Software AG and first installed in 1971.

Adabas is based on the NF² database model. NF² stands for NFNF = non first normal form. This term refers to the fact that Adabas does not store the data in the first normal form according to the relational theory. Instead, the data is stored in packed format that also allows periodic groups and multiple fields.

Adabas is used around the world together with the programming language Natural as high performance database. The database runs on the z/OS, z/VSE, Unix, Linux, and Windows. Other operating systems are also supported: BS1000, BS2000 (Siemens), VMS (DEC).

IDMS/DB

IDMS/DB (Integrated Database Management System) is based on a network model (CODASYL). Charles Bachmann (+ 2007), known for the so-called Bachman diagrams that are used for describing IDMS network topologies, participated in its development.

IDMS/DB was distributed by the American company Cullinet until 1989. In 1989, it was aquired by Computer Associates (CA / now Broadcom).

IDMS/DB was positioned as especially performant in comparison with relational databases. In the early to mid 80s relational databases became increasingly popular due to the ease of use, the query language SQL, and the emerging client server systems with better hardware systems.

Today, IDMS/DB is still used in countless installations around the world.

DATACOM/DB

DATACOM/DB was originally based on the concept of inverted lists, and is characterized by extremely fast analysis capabilities for large data volumes. The processing of large data volumes became much faster due to special index based technologies.

DATACOM/DB was aquired by Applied Data Research (ADR) in 1978. ADR also developed IDEAL as application language for DATACOM/DB.

In 1988 DATACOM/DB was taken over by Computer Associates (CA / now Broadcom). DATACOM/DB is very popular all over the world as database for mainframe users.

Db2

Db2 is IBM‘s relational database. It is based on the relational datamodel introduced by Edgar F. Codd in the early 70s.

DB2 (original spelling for Database 2) was introduced as name for the version of the operating system MVS by IBM in 1983.

The database was called SQL/DS for users of the operating systems VSE and VM. Now, only the official name Db2 is used, though VSE users may still call it SQL/DS.

Db2 is available on all IBM operating systems by now (z/OS, z/VSE, z/VM, AS/400) as well as on Linux, Unix, and Windows (Db2 LUW).

Db2 also uses VSAM as underlying data structure.

VSAM

Virtual Storage Access Method (VSAM) is an IBM DASD file storage access method, first used in the OS/VS1, OS/VS2 Release 1 (SVS) and Release 2 (MVS) operating systems, later used throughout the Multiple Virtual Storage (MVS) architecture and now in z/OS. Originally a record-oriented filesystem,[NB 2] VSAM comprises four[NB 2] data set organizations: Key Sequenced Data Set (KSDS), Relative Record Data Set (RRDS), Entry Sequenced Data Set (ESDS) and Linear Data Set (LDS).

VSAM was created as file system at the same time as the first virtual operating system OS/VS1. It is still one of the most important file systems for the mainframe operating systems z/OS and z/VSE.

There are different types of VSAM files:

  • KSDS: Key Sequenced Dataset (direct access via an index that is maintained in a separate index file. Sequential access is possible.)
  • ESDS: Entry Sequenced Dataset (access based on the sequential order in which the records were written)
  • RRDS: Relative Record Dataset (access based on the ordinal position in the file).
  • LDS: Linear Dataset (unstructured VSAM data set used by different system services).

Many system files and databases are based on VSAM (IMS/DLI, Db2, z/OS catalog, SMF, etc).

All databases and file systems of the operating systems z/OS and z/VSE listed above are supported by tcVISION as source and target.