Home / Publishers / Syncsort (Precisely) / MSU Optimization

Commercial desk · Syncsort (Precisely)

Syncsort (Precisely) Mainframe MSU Optimization.

Syncsort sort and integration tools ride on the same MSU capacity IBM bills. We move sort CPU to zIIP, reconcile the licensed capacity tier against the machines that actually run the software, and pull the baseline down before Precisely sets the renewal anchor.

48 hour mobilization

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

Get expert help →
№ 01

The situation

Capacity licensedzIIP offloadSCRT

Syncsort, now part of Precisely since the 2020 rebrand, licenses its mainframe estate mostly by capacity. Syncsort MFX, the high performance sort, copy, and join utility, is typically scaled to the MSU or MIPS rating of the machine or LPAR group it runs on, and Ironstream integration is sized to the feeds it moves. Because MFX is tied to the same capacity metric IBM bills, the cost moves with your general purpose MSU peak, which means the levers that reduce capacity based charges across the estate also move the Precisely number.

The lever that matters most here is the zIIP. Precisely's own ZPSaver states it can offload up to 90 percent of sort, copy, and SMS compression CPU cycles to the IBM zIIP engine, and zIIP work does not count toward the general purpose rolling four hour average that drives both IBM and capacity based vendor billing. Heavy sort workloads left on general purpose engines inflate the very peak your licenses ride on. The MSU optimization work exists to find and close that gap before renewal prices it back in.

№ 02

Our approach

BaselineReconcileLeverageClose

Five steps, in order

  • Baseline the capacity exposure. Which machines and LPARs the Syncsort software is licensed against, the capacity tier on each, and the general purpose MSU peak those workloads contribute, validated against your SCRT data.
  • Measure the sort profile. How much of the peak is sort, copy, and compression CPU that is a candidate for zIIP offload, and where ZPSaver or zIIP eligible paths would actually land given your workload mix.
  • Model the offload. The MSU the peak drops by once eligible sort CPU moves to zIIP, and the capacity tier the license could move to once the general purpose peak comes down. Offload is measured, never assumed.
  • Reconcile entitlement to footprint. Licensed capacity against the machines that actually run the software, with retired feeds and over scoped tiers flagged for the renewal.
  • Close with protections. The renewal repriced to the reduced baseline, capacity caps, and the right to re tier as the footprint changes, captured through contract review.
№ 03

A worked view of the lever

Illustrative

Directional only, to show the mechanic. Assume a peak hour where sort, copy, and compression contribute a meaningful share of the general purpose MSU peak that capacity based licenses ride on. Moving eligible sort CPU to zIIP removes it from the general purpose rolling four hour average.

Peak hour componentBefore offloadAfter zIIP offload
General purpose MSU peak1,200 MSU1,020 MSU
Of which sort / copy / compression200 MSU20 MSU
Eligible CPU moved to zIIP0up to ~90%
Peak the capacity license rides on1,200 MSU1,020 MSU

A 180 MSU reduction in the peak is a 15% cut to the baseline that both IBM MLC and capacity based vendors bill against, before any rate negotiation. Your actual eligible share and offload rate must be measured on your workload; the point is that the optimization is technical first and commercial second. See MSU consumption optimization for the full lever set.

№ 04

The directional outcome

Locked numbers

Across 500+ engagements and $180M+ of negotiated mainframe spend, disciplined MSU work on a proper runway typically produces renewal reductions of 20 to 35% against the initial quote, with the deepest cuts coming from a lowered capacity baseline rather than rate concessions alone.

$180M+

Mainframe spend negotiated

500+

Engagements delivered

20 to 35%

Typical renewal reduction

48h

Mobilization on audit notice or renewal

№ 05

Frequently asked

FAQ

How is Syncsort (Precisely) MFX licensed?

Syncsort MFX is commonly licensed by mainframe capacity, scaled to the MSU or MIPS rating of the machine or LPAR group it runs on. That ties the cost to the same capacity metric IBM bills, so the same MSU baseline discipline that cuts IBM charges also moves the Precisely number.

Does ZPSaver actually reduce our MSU baseline?

Precisely states ZPSaver can offload up to 90 percent of sort, copy, and SMS compression CPU cycles to the zIIP processor. zIIP work does not count toward the general purpose MSU rolling four hour average, so heavy sort workloads moved to zIIP can lower the peak that both IBM and capacity based vendors bill against. The actual offload depends on your workload mix and must be measured.

Where do the largest Syncsort (Precisely) savings usually come from?

In our engagements the largest reductions typically come from two levers: validating that the licensed capacity tier matches the machines the software actually runs on, and offloading sort CPU to zIIP so the general purpose peak the license rides on comes down. Retiring Ironstream feeds no longer consumed is a frequent third.

Can we optimize MSU exposure mid term, before renewal?

Yes, and you typically should. zIIP offload, capacity reconciliation, and SCRT validation build the evidence base the renewal negotiation later runs on, and the MSU reduction starts compounding immediately. We mobilize within 48 hours.

Related desks: Syncsort (Precisely) license negotiation, Syncsort (Precisely) cost optimization, and the firm wide MSU optimization service.

The peak sets the price. Bring it down.

Get expert help