Prepare
- Upload CA IDMS files
Guide
IDMS is a network database management system based on the CODASYL standard, commonly used in mainframe enterprise environments.
Why Agent Required
IDMS runs on mainframe systems that cannot directly upload data to the web. The DataMeans Agent connects securely to your mainframe environment for extraction.
What You Can Extract
- IDMS versions 10.x through current
- Database records and structures
- CODASYL set relationships
- Multi-valued fields
What You Get Out
DataMeans extracts your data into multiple modern formats:
| Output | Description |
|---|---|
csv/{TableName}.csv | One CSV file per table with all row data |
xlsx/{TableName}.xlsx | Excel workbook per table |
xls/{TableName}.xls | Legacy Excel format per table |
json/{TableName}.json | JSON array of records per table |
json/{TableName}.jsonl | Newline-delimited JSON (streaming-friendly) |
postgres.sql | PostgreSQL CREATE TABLE + INSERT statements |
schema/schema-graph.json | Relationship graph for visualization |
schema/er-model.json | ER model for diagram tools |
report.json | Structured extraction report |
report.md | Human-readable extraction summary |
How It Works
Agent availability: The DataMeans Agent is currently in development and not yet available for download. The steps below describe the planned workflow. To get notified when IDMS support launches, reach out via our About page.
- Download and install the DataMeans Agent on a system with IDMS access
- Configure connection to your mainframe/IDMS environment
- Select schemas and tables to extract
- The Agent extracts data and uploads securely to DataMeans
- Review and download your converted data
Planned Support
- Record structure extraction
- CODASYL set relationship mapping
- Pointer chain resolution
- Multi-valued field normalization
Known Limitations
- Requires DataMeans Agent installation
- Mainframe connectivity required
- Network database model requires flattening to relational format
Last updated: January 2026
Technical reference
Overview
CA IDMS (Integrated Database Management System) is a network database management system based on the CODASYL standard, designed for IBM mainframe environments. It uses a network model where data is organized in records connected through sets, with owner-member relationships allowing complex navigational access patterns. Unlike hierarchical databases, IDMS supports many-to-many relationships and flexible data structures, optimized for high-volume transaction processing in enterprise mainframe systems.
History and Background
- 1964: Integrated Data Store (IDS), developed at General Electric by a team led by Charles Bachman, is first released; it is the direct ancestor of IDMS.
- 1969-1971: B.F. Goodrich Chemical Division rewrites GE's IDS for IBM System/360 mainframes, naming the result IDMS.
- 1973: Charles Bachman receives the ACM Turing Award for "his outstanding contributions to database technology".
- 1973: Cullinane Corporation acquires the rights and begins selling an enhanced IDMS, paying royalties to B.F. Goodrich.
- 1978: Company renamed Cullinane Database Systems prior to going public.
- 1979: The IDMS-DC teleprocessing monitor is marketed alongside the database, adding online transaction processing.
- 1982: Cullinane Database Systems becomes the first computer software firm listed on the New York Stock Exchange.
- 1983: Company renamed Cullinet Software; IDMS/R (Release 10.0) is unveiled, adding relational-style features to the network core.
- Mid-1980s: Competing with IBM's DB2 and other relational databases, Cullinet develops a relational front-end and productivity tools, including the Automatic System Facility (ASF) built on the existing Logical Record Facility (LRF).
- 1989: Computer Associates acquires Cullinet; the product is later branded CA-IDMS.
- 1992: Release 12.0 delivers full SQL support, catalog management, and a centralized security facility.
- 2008: Release 17.0 introduces support for IBM zIIP specialty processors.
- 2018: Broadcom acquires CA Technologies and continues development under the IDMS name.
File Format Specifications
IDMS uses proprietary file formats optimized for mainframe storage systems.
File Extensions:
- No specific extensions - uses standard mainframe datasets
- Database files: VSAM ESDS clusters or non-VSAM files accessed via BDAM or EXCP
- Journal files: For transaction logging and recovery
- Dictionary files: Schema metadata held in the Integrated Data Dictionary (IDD)
File Structure:
- Areas: Pre-formatted subdivisions of database files
- Pages: Subdivisions of areas corresponding to physical disk blocks
- Page header: First 16 bytes of each page — a 4-byte page number plus a 12-byte SR1 system record holding the page's free-space count
- Page footer: Last 16 bytes — line index 0, a line space count, filler, and a repeated 4-byte page number
- Line indexes: One 8-byte entry per stored record (record id, displacement, record length, and prefix length, 2 bytes each); records fill pages top-down while line indexes build bottom-up, leaving free space in the middle
- Page initialization: The FORMAT utility initializes pages, setting usable space to the page size minus the 32 bytes of header and footer
- Database keys (db-keys): 4-byte binary values concatenating a page number and a line number
- Default db-key format: 24 bits for the page number and 8 for the line number, allowing 16,777,214 pages of up to 255 records each; the line-number portion can range from 2 to 12 bits
- Records: Data entities following COBOL field conventions
- Sets: Owner-member relationships maintained as linked lists
- Record placement: DIRECT, SEQUENTIAL, CALC (hashed), and VIA (clustered near owner) modes
Key Components:
- Schema: Database structure definition
- Subschema: Application view of schema
- DMCL: Device Media Control Language
- IDD (Integrated Data Dictionary): Maintains database definitions and also stores definitions and code for companion products such as ADS/Online and IDMS-DC
- Journaling: Transaction recovery mechanisms
- Security: Centralized security facility (Release 12.0)
Data Types and Structures
Record elements follow COBOL conventions, supporting repeating items and group fields.
| Type | Description | Storage |
|---|---|---|
| CHARACTER | EBCDIC text strings (USAGE DISPLAY) | 1 byte per character, up to 32,767 bytes |
| ZONED DECIMAL | Decimal numbers (USAGE DISPLAY) | 1 byte per digit, 1-18 digits |
| PACKED DECIMAL | Decimal numbers (COMP-3) | Half byte per digit, 1-18 digits |
| BINARY | Integers (COMP) | Halfword (2), fullword (4), or doubleword (8 bytes) |
| FLOATING POINT | Approximate numerics (COMP-1/COMP-2) | 4 bytes (short) or 8 bytes (long) |
| MULTIBIT BINARY | Bit strings (USAGE BIT) | 1 bit per position, 1 to 32 bits |
| GROUP | Container of subordinate data fields | No PICTURE or USAGE clause of its own |
The SQL option adds types including SMALLINT, INTEGER, BIGINT (LONGINT), DECIMAL, NUMERIC, REAL, DOUBLE PRECISION, FLOAT, CHARACTER, VARCHAR, GRAPHIC, VARGRAPHIC, BINARY, DATE, TIME, TIMESTAMP, ROWID, and XML.
Network Model:
- Records: Primary data containers
- Sets: Define relationships between records
- Owners: Parent records in sets
- Members: Child records in sets
- Pointers: Database-key links forming chains (next, optional prior and owner)
- Set membership: Declared MANDATORY or OPTIONAL and AUTOMATIC or MANUAL, governing CONNECT and DISCONNECT behavior
- Set order: New members are positioned FIRST (LIFO), LAST (FIFO), NEXT, PRIOR, or SORTED by sort key
Version Differences
| Version | Year | Key Changes | Compatibility |
|---|---|---|---|
| Release 12.0 | 1992 | Full SQL support, centralized security, catalog and deadlock management | Logical/physical database separation |
| Release 14.0 | 1999 | Multitasking support | Parallel Sysplex exploitation |
| Release 15.0 | 2001 | Improved performance | Data sharing support |
| Release 16.0 | 2004 | Two-phase commit, XML publishing, Parallel Access Volumes exploitation | TCP/IP support |
| Release 17.0 | 2008 | Improved automatic recovery | zIIP processor support |
| Release 18.0 | 2011 | Automatic system tuning, simplified installation and maintenance | Enhanced zIIP exploitation |
Compatibility Notes:
- SQL access added in Release 12.0 alongside the original navigational DML
- Database files accessed through VSAM, BDAM, or EXCP access methods
- Native VSAM files (KSDS, ESDS, RRDS) can be accessed without conversion
- Runs on IBM z/OS and z/VSE operating systems
- IDMS Server provides distributed access through ODBC and JDBC drivers supporting dynamic SQL; the JDBC driver conforms to the JDBC 4.0 specification
- ODBC and JDBC driver traffic reaches the central version over TCP/IP
- An MQ Adapter (added to 19.0 in 2019) lets IDMS applications exchange messages with IBM MQ queue managers
Technical References
- Broadcom IDMS 19.0 Documentation
- IDMS SQL Reference: Data Types
- Wikipedia: IDMS
- Wikipedia: Cullinet
- Wikipedia: CODASYL
To learn how to use this format with DataMeans, see the User Guide.