Publishers · IBM · MSU optimization

IBM Mainframe MSU Optimization.

Under classic MLC pricing, one four hour window a month sets the bill for z/OS, CICS, Db2, IMS and MQ. Under Tailored Fit Pricing, every MSU you consume counts against a negotiated baseline. Either way, the cheapest MSU is the one you stop paying for, and most estates carry more of them than anyone has measured.

The bill is made in four hours.

IBM's Monthly License Charge products are billed on the rolling four hour average, the R4HA, reported through SCRT. The bill does not care about your monthly average; it cares about the single highest four hour window across the month, per LPAR, per product. A batch collision at 2 a.m. on one Tuesday can set the price of the entire month. Estates that have never profiled their peak typically discover it is shaped by scheduling accidents, not business demand.

Under the Tailored Fit Pricing Software Consumption Solution the mechanics change but the stakes do not: consumption is billed annually against a baseline IBM derives from your trailing SCRT history. Accept a baseline set on an unoptimized estate and you lock the waste in. Optimize first, then negotiate the baseline, and the savings compound for the life of the agreement.

Our approach · five steps

1 Step 01

Profile the peak

Twelve months of SCRT and SMF data, mapped hour by hour. We identify exactly which workloads create the R4HA peak on each LPAR, when, and why. This map is the foundation; everything else is executed against it.

2 Step 02

Move what makes the peak

Batch window restructuring and peak shaving: rescheduling discretionary work out of the peak window, separating workloads that collide, and aligning WLM policy so low importance work yields when the R4HA is forming.

3 Step 03

Offload what qualifies

Eligible work moved to zIIP engines, where it consumes no billable MSU. Db2 DRDA, utilities, and growing classes of Java and system work qualify; estates commonly run zIIP eligible work on general processors simply because nobody checked.

4 Step 04

Cap what remains

Defined capacity and group capacity limits, modeled against the real workload profile so the cap cuts billing peaks without throttling the business. Capping an unprofiled estate is guesswork; capping a profiled one is engineering.

5 Step 05

Reset the commercial terms

The optimized consumption profile becomes the negotiating position: a corrected TFP baseline, right sized MLC, and contract language that captures the gains permanently. Technical savings that never reach the contract typically leak back within two renewal cycles.

What changes with us on the estate.

Your team keeps running the platform; we bring the measurement discipline and the commercial follow through. IBM's own tooling reports what you consumed. It does not volunteer where consumption was avoidable, and the account team is not incentivized to find out. We are. Across 500+ engagements and $180M+ of negotiated mainframe spend, the pattern holds: the estates that measure their peaks pay less than the estates that accept them.

What you get

  • An hour by hour map of what creates your R4HA peak, per LPAR, per product
  • A ranked savings plan: scheduling moves, zIIP offload, capping strategy, each sized in MSU and dollars
  • An independently validated SCRT position you can defend in any IBM review
  • A TFP baseline or MLC position negotiated on the optimized number, not the historical one

Frequently asked

Q1

How much can MSU optimization save?

It depends on how much of the bill is driven by avoidable peaks and ineligible workload placement, so we do not quote a universal percentage. Across 500+ engagements, the combination of consumption work and negotiation typically produces 20 to 35% reductions at renewal, and MSU work is frequently the largest single contributor.

Q2

Does it require changing applications?

Usually not. The biggest levers are operational: when batch runs, how WLM policies shape the peak, what runs on zIIP, and how defined capacity is set. Application changes are a later, optional lever once the operational gains are captured.

Q3

We are on Tailored Fit Pricing. Still relevant?

More than ever. Under the Software Consumption Solution every MSU consumed is billable against your baseline, not just the monthly peak. Consumption discipline cuts the annual bill directly, and the baseline itself is negotiable at entry and at renewal.

Q4

Is soft capping safe for production?

Done correctly, yes. Defined capacity and group capacity limits are IBM supported mechanisms. The craft is in setting limits that cut billing peaks without throttling business critical work, which is why we model the R4HA profile before recommending any cap.

Q5

What data do you need to start?

Twelve months of SCRT reports, SMF type 70 and 89 records if available, your MLC invoices, and the current contract set. With that we can typically show where the bill is made, and where it can be unmade, within the first weeks.

Related

The IBM buyer side guide →
01 IBM cost optimization

The full cost programEntitlements, shelfware and contract structure beyond the MSU line. IBM mainframe cost optimization →

02 IBM renewal advisory

Carry the gains into the contractThe renewal program that locks optimized consumption into the terms. IBM mainframe renewal advisory →

03 The technical levers

MSU consumption optimizationThe explainer behind this service: every lever, with worked examples. Read the explainer →

Your peak is setting the price. Find out why.

Get expert help

Audit notice or renewal under 18 months out? We mobilize within 48 hours.

Get expert help