① Commercial · Syncsort (Precisely)
Syncsort MFX is one of the most widely installed third party products on the mainframe, and it renews on capacity. When you upgrade the box, the sort bill climbs even if your workload did not. We optimize the capacity basis, the zIIP offload, and the Ironstream volume tier before the renewal locks them in.
Capacity based pricing means a hardware refresh is also a price increase you never agreed to.
Syncsort, rebranded under Precisely after the 2020 merger, sells a focused mainframe portfolio: Syncsort MFX for high performance sort, copy, and join, the ZPSaver feature that offloads most of that work to zIIP, and Ironstream for streaming operational data into platforms such as Splunk and ServiceNow. MFX is typically licensed on machine capacity, so the cost driver is the size of the box, not the value of the work. A capacity upgrade or a consolidation onto a larger machine raises the MFX bill automatically, and that increase is rarely questioned because it arrives as a renewal rather than a quote.
Cost optimization here is mechanical, not adversarial theater. The capacity basis can be validated against actual MFX usage, ZPSaver can be enabled or right sized to move work onto zIIP engines that fall outside general purpose software charges, and Ironstream can be tiered to the data volume you actually feed rather than the forecast you committed to. Where leverage is needed, IBM DFSORT ships with z/OS and provides a credible, already installed alternative for the sort function. The publisher level picture is on the Syncsort (Precisely) hub; the zIIP mechanics are explained in zIIP engines and software cost offload.
Confirm what capacity MFX is actually licensed against and whether it tracks real sort workload. Where the basis was set on an old machine model or a peak that no longer applies, the renewal is the moment to correct it.
ZPSaver can offload the large majority of sort, copy, and compression cycles to zIIP engines. We confirm it is enabled, sized correctly, and actually reducing general purpose CPU, because that consumption drives charges across your whole software stack, not just MFX.
Ironstream tiers are commonly overcommitted at signing against a forecast nobody measured. We size the tier to observed data volume and negotiate headroom for growth rather than paying for a high fixed floor from day one.
IBM DFSORT is already on the floor with no incremental sort license. We cost the switch honestly, including JCL and parameter migration, so the walk away is evidenced. The point is rarely to switch; it is to make Precisely price the renewal as if you might.
Capacity protections, escalator ceilings, and volume terms negotiated while leverage exists, on a sequence that starts 12 to 18 months out. When the engagement starts late, we compress: a renewal under 18 months out mobilizes within 48 hours.
Mainframe spend negotiated across the major publishers
Engagements delivered since 2019
Typical renewal reduction against the vendor's opening position
Audit notice or renewal under 18 months out? We mobilize within 48 hours. Upgrading the box and watching the Syncsort bill climb? Optimize the basis before it renews.
Syncsort MFX (Precisely) is typically licensed on mainframe capacity, measured in MSU or by machine model, so the bill rises when you upgrade hardware even if your sort workload is flat. ZPSaver is a separately licensable MFX feature that offloads most sort, copy, and compression work to zIIP engines, which can reduce general purpose CPU consumption and the costs tied to it. Product detail sits on the Ironstream licensing page.
Usually yes. The fastest wins come from validating the capacity basis against what MFX actually uses, enabling or right sizing ZPSaver to push work onto zIIP, and tiering Ironstream by real data volume rather than a high committed forecast. Replacement, typically a move back to or away from IBM DFSORT, is a leverage option, not the only path to savings.
IBM DFSORT ships with z/OS, so the credible alternative for the sort function is already on the floor. MFX is widely retained for performance and zIIP offload, but the existence of a no incremental license alternative is real leverage at renewal. We cost the switch honestly, including JCL and parameter migration effort, so the walk away is evidenced rather than asserted.
Ironstream, which streams mainframe and IBM i operational data into platforms such as Splunk and ServiceNow, is commonly priced on data volume or connected capacity. Overcommitting the volume tier at signing is a frequent source of overpayment, because the forecast is set before anyone measures the real feed. We size it to observed volume and negotiate growth room rather than a high fixed floor.