IDMS: the database that renews on capacity, not technology.

IDMS is Broadcom's (CA) established mainframe database, a network model database with a relational interface that still runs mission critical systems in banking, insurance, government, and utilities. Broadcom acquired it as part of the CA Technologies portfolio, the largest non IBM mainframe software estate. The database itself is rarely the reason a shop keeps it; the reason is the application layer above, often decades of embedded business logic and integrations that would take a multi year program to move. That dependency is the central fact of IDMS licensing, because it limits the credible walk away.

IDMS is licensed on processor capacity, historically expressed in MIPS and increasingly in MSU as Broadcom (CA) aligns the portfolio to the MSU metric. The charge tracks the capacity of the environment IDMS is authorized to run on, typically the LPARs or machines named in the agreement. On a Broadcom Mainframe Consumption Licensing (MCL) arrangement the model shifts from a fixed per machine entitlement toward measured MSU consumption against a committed baseline, reconciled annually through True Forward. Many estates carry IDMS inside a wider Broadcom (CA) bundle, so its cost is frequently set by the portfolio renewal rather than a standalone IDMS line.

The first cost driver is the licensed capacity or the committed MSU baseline, which on a stable IDMS estate often sits above true consumption because it was set in a busier era and never reset. The second is the renewal uplift: Broadcom (CA) renewals commonly carry annual escalators, so even a flat IDMS workload can see the bill climb year over year unless the uplift is capped. The third is the bundle effect, where growth in other products under the same agreement lifts a shared consumption commitment that IDMS is part of, so IDMS cost rises for reasons that have nothing to do with IDMS.

IDMS exposure usually sits in the gap between what the contract assumes and what the estate actually runs. Common traps we see at pattern level:

With displacement off the table for most estates, the renewal is where IDMS cost is actually won or lost. The five levers that pay:

Leaving IDMS means rewriting or migrating the applications built on it, which is a multi year modernization program with real risk, not a renewal tactic. That said, the credible pressure is genuine where it exists: organizations do migrate IDMS workloads to other databases as part of broader modernization, and even a credible, costed plan to do so over time strengthens the negotiating position today. Third party support is another lever some estates use on stable, low change IDMS environments. Treat any pitch that promises a quick, low risk IDMS exit with caution, because the application dependency, not the database, is the hard part.