Today’s article in my series on zIIP processors will be a little different because I want to focus on a specific IBM mainframe pricing model and how users of that model may benefit from using zIIPs. The pricing model, Tailored Fit Pricing (TFP), is the latest model introduced by IBM. But what is TFP, and how does it differ from the sub-capacity pricing models used by most organizations?
The Current State: Sub-Capacity Pricing Models
IBM touts TFP as a flexible pricing model that is simpler and more cloud-like than other pricing options. And that is true, but you really need to understand the particulars of TFP to ensure that it makes sense for your organization.
Today, most mainframe sites use some form of sub-capacity pricing model that relies on measuring usage and billing based on the peak, monthly Rolling 4 Hour Average (R4HA) of MSU consumption. This may include Variable Workload License Charges (VWLC), Advanced Workload License Charges (AWLC), System z New Application License Charges (zNALC), and others. All of these sub-capacity pricing models require you to use the Sub-Capacity Reporting Tool (SCRT) to create reports that track peak R4HA usage by product and LPAR and then submit those reports to IBM for billing purposes.
Based on those reports, your organization pays IBM monthly and your payment may vary based on your monthly peak usage. The primary benefit of sub-capacity pricing models is that your payment amount somewhat corresponds to your use. However, many organizations want a more consistent monthly bill because it can be difficult to cost-plan when usage fluctuates monthly.
Additionally, many IBM z customers attempt to tune their applications based on the peak R4HA to reduce costs. However, this can be very difficult to achieve as different combinations of products and LPARs can peak on different days and at different times. So they take intricate steps like soft-capping, placing restrictions on test usage, and even diverting new workloads to other platforms even when the mainframe is the most sensible deployment environment.
What is Tailored-Fit Pricing?
This brings us to IBM’s tailored fit pricing model. The general idea of TFP is to enable a more predictable cloud-like pricing option for IBM z software. There are several TFP options, including:
• Software Consumption Solution,
• Hardware Consumption Solution,
• Development and Test Solution,
• Enterprise Capacity Solution, and
• New Application Solution.
We will not get into all the details of each solution but will offer a short introduction.
The TFP Software Consumption model is designed to accommodate modern workloads typified by being spiky. That means the workload is not a consistent rolling one but has peaks and valleys that vary widely over time. This model also removes the need to cap workload as it is designed to enable a consistent, regular bill based on past workloads and usage. The client and IBM work together to create a baseline usage calculated from the past 12 months of MSU consumption. This is combined with a negotiated price per MSU which is rolled together to create a predictable bill for the year. Furthermore, any growth qualifies for a reduced price per MSU.
The TFP Hardware Consumption model enables customers to combine a base capacity with a subscription-based capacity as needed, which can be applied for usage spikes. The usage charges have a granularity of 1 hour and are based on actual the MSUs consumed as measured by the SCRT, not full engine capacity.
The TFP Development and Test allows customers to use workload more flexibly for development and testing. The workload must operate on separate LPARs. The goal is for developers to always have sufficient capacity for development and testing at a specific cost per MSU.
The TFP Enterprise Capacity option is a full-capacity licensing model that delivers a predictable monthly cost. Software charges are based on the overall size of your physical hardware environment. Charges are calculated based on the estimated mix of workloads running while providing the flexibility to vary actual usage across workloads.
And finally, the TFP New Application solution is designed to enable new workloads to be introduced to the IBM z at a reasonable price point.
The general idea is that TFP is designed to make billing more straightforward and consistent. It also has built-in plans for growing workloads and pricing them accordingly.
The Impact of TFP on zIIPs (and vice versa)
With TFP, every MSU saved can result in cost savings when your baseline is calculated for the following year. Note that there are no immediate cost savings as your monthly software bill is agreed upon at the beginning of your contract year based on last year’s baseline usage. So, if you save an MSU, any MSU, it can impact next year’s bill. Also, we are not talking about MSUs in a peak R4HA, as that is not relevant for most TFP charges.
With this in mind, you should consider how zIIPs can reduce costs. Remember that workload running on zIIPs is not chargeable. For TFP, the workload running on zIIPs will not factor into the calculation of your baseline for next year.
Using the methods we have described in past articles, you can redirect workload to zIIPs, reducing your overall MSU consumption. A sub-capacity pricing model may or may not reduce your cost depending on whether the workload ran during an R4HA peak. With TFP, any MSUs redirected to the zIIP will reduce your MSU baseline for next year’s pricing.
The Bottom Line
More and more organizations are evaluating the impact of tailored fit pricing and moving away from sub-capacity models to improve the predictability and transparency of their monthly IBM z software bill. As such, wise organizations will simultaneously look to bolster usage of their zIIP processors to reduce their costs further as they move to TFP.
Craig Mullins, Copyright Craig Mullins Consulting, Inc
Find Out More
Try our Business Case Calculator.