① Journal · Benchmarks
With Broadcom (CA), the benchmark that matters is not the rate per MIPS. It is how much of the bundle you actually use. Across the CA mainframe renewals we work, one or two families carry real workload while a large share of the portfolio sits idle but fully paid. Benchmarking Broadcom spend means counting that idle before the renewal letter arrives and locks it in for another term.
48 hour mobilization Audit notice or renewal under 18 months out? We mobilize within 48 hours.
Get expert help →Broadcom (CA) prices its mainframe portfolio on capacity, historically in MIPS and increasingly converted to MSU under Mainframe Consumption Licensing, with an annual True Forward adjustment that trues the committed capacity upward as the estate grows. Large enterprises commonly run twenty to forty distinct CA product families across systems management, scheduling, security, database, storage, and DevOps tooling. The bundle structure is the trap: families travel together, the capacity figure covers all of them, and a product nobody has opened in years is billed at exactly the same capacity as one running every batch window.
That is why a per MIPS rate benchmark misleads. Two shops can pay an identical rate and sit worlds apart on value, because one runs most of what it licenses and the other carries a portfolio where a meaningful share is dormant. The benchmark that predicts savings is utilization by family, not the headline rate.
Across roughly twenty to thirty Broadcom renewals, the bill divides into a small core that earns its place and a long tail that does not. The benchmark exercise is to tag every family against honest usage, not against the entitlement list:
| Portfolio segment | Typical state | Benchmark signal |
|---|---|---|
| Core in use (scheduling, security, key utilities) | Real daily workload, defensible | Benchmark the capacity, not whether to keep it |
| Partially used families | One module live, the rest dormant | Candidate to unbundle or downsize at renewal |
| Idle but billed | Installed, entitled, not opened in years | The 30 to 50 percent that the renewal quietly re-locks |
| Capacity creep | True Forward trues the whole bundle up with the estate | Check the cap, not just the rate |
When capacity grows without a negotiated cap, renewal uplifts in the range of 30 to 80 percent are commonly observed, because the True Forward applies across the full bundle, idle families included. The single most effective lever is rarely the rate. It is retiring or unbundling the families that no longer earn their place before they are re-committed for another term.
A defensible Broadcom benchmark starts with a usage inventory taken from the estate, not from the contract. For each family, record whether it ran in the last measured period and who depends on it. Express the bill as cost carried by idle capacity versus cost carried by working capacity, because that split is the number that moves a negotiation. Then anchor the capacity commitment and cap the True Forward, so growth in the estate does not silently re-inflate the dormant tail every year. Benchmark bands from the wider market are a sanity check on the rate, but the rate is the smaller half of the story here. Utilization is the larger one.
The method generalizes in the guide on benchmarking your mainframe software spend, and the renewal mechanic itself is covered in the renewal uplift letter decoded. The sibling read on the other half of most bills is benchmarking IBM mainframe spend. When the Broadcom renewal arrives and the True Forward is on the table, our Broadcom (CA) MSU optimization work turns the utilization count into a concrete reduction. The estate level view sits in the Broadcom (CA) licensing hub.