① Licensing concept · MLC metrics
One metric, six labels, twenty five years of IBM Z pricing history. Knowing which WLC variant your contract uses, and how its MSU curve tiers, decides what every MSU you shave is actually worth.
The sub-capacity metric that pays you to manage the peak. Six variants, one curve.
Workload License Charges, WLC, are IBM's Monthly License Charge metric for z/OS software, introduced in 2000 with the zSeries 900 and the z/OS operating system. WLC made sub-capacity charging possible: eligible products are billed on the measured rolling four hour average MSU peak of the LPARs that run them, not on the full machine. The major z/OS programs are WLC eligible, including z/OS, CICS, Db2 for z/OS, IMS and WebSphere MQ, which is why WLC sits under most large MLC bills.
Over the years IBM layered variants on top, mostly to align pricing with new machine generations. VWLC is the base variable metric. AWLC improved price performance on the higher end servers, and AEWLC did the same at the entry end, building on EWLC. FWLC covers certain legacy products at a flat rate. The labels matter because the variable ones tier down, rewarding peak management, while the flat one does not. Reading which variant each product carries is the first step in knowing where cost actually moves.
Eligibility and applicability depend on machine generation and the product. Confirm which variant each program carries against your own contract and SCRT output.
| Variant | What it is | Buyer relevance |
|---|---|---|
| VWLC · Variable WLC | The base variable sub-capacity metric, cost per MSU falls as capacity rises | Peak management directly lowers the bill |
| AWLC · Advanced WLC | Variable metric aligned to later high end machine generations | Better price performance on eligible high end servers |
| EWLC · Entry WLC | The variable entry metric for lower end machines | Applies to smaller estates and machines |
| AEWLC · Advanced Entry WLC | The advanced entry variant for newer lower end servers | Entry equivalent of AWLC price performance |
| FWLC · Flat WLC | Flat rate per MSU on certain legacy products | Does not tier down, so capping does not cut its rate |
| Defined Capacity and GCL | The capping controls that set the billable MSU under variable WLC | The levers that move the rolling four hour average |
Variable WLC prices MSU in graduated bands: each higher band costs less per MSU than the one below it. The figures here are illustrative band rates, expressed as rate units U per MSU, to show the mechanic. Actual rates are negotiated and are not stated here.
| MSU band | Illustrative rate per MSU | What this means |
|---|---|---|
| Band 1 · first 45 MSU | 10 U | The most expensive MSU on the machine |
| Band 2 · next 130 MSU | 7 U | Cheaper per MSU than band 1 |
| Band 3 · next 400 MSU | 5 U | Cheaper again |
| Band 4 · above 575 MSU | 3 U | The cheapest MSU, at the top of the curve |
Because the bands fill from the bottom up, the MSU you remove by capping or scheduling come off the top band you currently reach. Shaving a peak from 620 down to 560 MSU removes MSU priced near the cheapest band, so the saving is modest. Shaving from 180 down to 120 removes MSU from a more expensive band, so the same 60 MSU saves more. Knowing where your billed peak sits on the curve tells you whether peak shaving is worth the engineering effort, or whether the bigger lever is renegotiating the rate.
FWLC products charge a flat rate per MSU regardless of band, so the operational discipline that cuts variable WLC does nothing for them. Treating a flat product like a variable one wastes effort on the wrong lever.
A machine refresh can make a better variant available, but the contract does not update itself. Estates commonly keep paying an older variant after a hardware change that entitled them to advanced price performance.
Under variable WLC a single uncontrolled month sets the bill. Without group capacity limits and defined capacity in place, one spike resets the rolling four hour average upward for that month's charge.
The same MSU reduction is worth more at the bottom of the curve than the top. Optimization that ignores where the billed peak sits on the band structure can spend engineering effort for little return.
Know the variant, read the curve, then pick the lever.
Start by mapping which WLC variant each program carries and where your billed peak sits on the band curve, because that determines whether the cheaper lever is operational or contractual. For variable products high on the curve, capping and peak shaving work, and the controls are defined capacity and group capacity limits. For flat products and for products already low on the curve, the lever is the rate and the contract, not the engineering. After a hardware refresh, check whether a better variant became available and was actually applied.
The metric underneath is the MSU, so read MSU explained and the rolling four hour average for how the billed figure is built, and the pricing history from PSLC to TFP for where WLC sits in the timeline. When a renewal puts WLC rates and variants in play, our mainframe license negotiation and cost optimization teams read the curve against your own data before you agree a number.
IBM's Monthly License Charge metric for z/OS software, introduced in 2000, that enables sub-capacity billing on the rolling four hour average MSU peak. The major z/OS programs including CICS, Db2 for z/OS, IMS and MQ are WLC eligible.
VWLC is the base variable metric. AWLC gives better price performance on later high end machines, and EWLC and AEWLC are the entry equivalents. FWLC is a flat rate on legacy products that does not tier down.
Variable WLC prices MSU in bands where each higher band costs less per MSU. Reductions come off the top band you reach, so the same MSU saved is worth more lower on the curve than higher.
No. WLC bills on the monthly four hour peak. Tailored Fit Pricing, from 2019, bills against a negotiated annual baseline. The two reward opposite behavior, so estates model both before moving eligible workloads.