Broadcom CA IDMS · from zero

The mainframe database you have to copy data out of, explained from scratch.

CA IDMS stores data in a way that feels alien if you have only used normal databases. It does not join tables by matching values — it follows physical addresses written onto the records themselves. That one difference explains everything else, and it is why copying the data somewhere modern is harder than it looks.

This course assumes no mainframe knowledge at all and takes you to the point where you can read one of these databases, write code against it, and design the system that replicates it.

Start here — no experience needed → Skip the primer, go to Module 01 Glossary

0 units complete (0%)

11units, ~9 hours
13interactive simulators
1live database interpreter
50question final exam
Never seen a mainframe?

Good — that is exactly who this was written for. Nothing here assumes you have used one, and no term is used before it is explained.

Words that might be unfamiliar are underlined with dots throughout the course. Hover over one, or tap it on a phone, and a one-sentence explanation appears without leaving the page. Start with the primer — twenty minutes, and everything after it gets easier.

Who this is for

You build software. You know relational databases. You have just been told that a system you need to work with runs on a mainframe, and the documents you were handed are full of words like set, area, currency and db-key that nobody has defined for you.

This course closes that gap properly. Not with a summary — with the actual mechanics, taught in plain language, so that when a data load fails or a copy silently stops matching, you know where to look and why.

You should already know

  • Databases and SQL, roughly
  • That transactions and logs exist
  • How to read code in some language

You do not need

  • Any mainframe experience
  • Access to a real IDMS system
  • COBOL — we explain what matters
  • To have heard of CODASYL

You will leave able to

  • Read a database design and predict how it behaves
  • Write and debug code against it
  • Design a sensible target database
  • Diagnose a broken data pipeline on evidence

The one idea the whole course rests on

In a normal database, a relationship is a value. You put 3200 in an employee row and 3200 in a department row, and the database matches them up when you ask.

In IDMS, a relationship is an address. Each record physically carries the location of the next related record, like a linked list stored on disk. Nothing is matched. You go where the record tells you to go.

Picture it

A warehouse of numbered shelves. Each shelf has numbered slots. Each slot holds one index card.

A relationship is a labelled loop of string tying a parent card to each of its children and back again. To find a department's employees you do not search — you pick up the department card and follow its string.

The whole course is details of that warehouse: how shelves are numbered, how you decide which shelf a card goes on, what happens when a shelf fills up, and what the string costs you when things change.

Why you are probably here Every module ends with a section called What this means for replication — meaning, what it affects when you copy this data to a modern system using Rocket Data Replicate and Sync. Read the module to understand the database; read that section for the reason it landed on your desk. Module 10 assembles all of them into the real architecture.

The course

How to work through it

  • Start with the primer if any of this is new. Twenty minutes, and everything after it gets easier.
  • Do the modules in order the first time. Each builds on the last, and Module 10 assumes all nine before it.
  • Use the simulators. They are not decoration — the one in Module 05 is a real working database you type commands into. Some things only become obvious by doing them.
  • Hover the dotted words instead of trying to memorise vocabulary. It sticks by use.
  • Progress is saved in this browser only. Nothing is uploaded anywhere.
  • Keyboard: j next unit, k previous, Esc closes the menu.
  • Checkpoints end every module, scored, with an explanation for every answer. 75% is the bar.

One honest caveat

The names, commands, design syntax, storage concepts and architecture in this course are faithful to the real product. The sample data, page numbers, the storage-placement formula and the performance models are teaching constructions — the same shape as the real thing, not exact reproductions of any particular release.

Before running anything against a real system, check it against your site's own manuals for your version. Details do change between releases, and this course cannot track them.