Rocket Host Access licensing: a seat priced emulator in mid consolidation.

Rocket Host Access is Rocket Software's terminal emulation line, the software that gives desktops secure 3270, 5250, and VT sessions to mainframe, IBM i, and UNIX hosts so that green screen applications stay reachable from a modern, managed client. RUMBA, originally developed by NetManage and long popular across AS/400, iSeries, and IBM i shops, reached Rocket through its 2024 acquisition of a large part of the former Micro Focus portfolio from OpenText. With that deal Rocket inherited roughly six emulators, RUMBA, HostExplorer, EXTRA!, Reflection, BlueZone, and Passport among them, and has been consolidating that overlapping estate under Rocket Secure Host Access, a security first emulator, while continuing to support the established lines. Host Access is therefore best understood as a family in mid consolidation rather than one fixed product, which is precisely the condition that shapes its licensing.

Rocket Host Access and RUMBA are licensed per seat, by named or concurrent user, scaled to the number of desktops that run an emulator session, rather than by mainframe MSU capacity. Historically these lines were sold as perpetual licenses with annual maintenance. As Rocket consolidates the emulator portfolio it has been moving toward subscription, and has offered migration help, including bundled consulting hours, to move customers off any of the legacy emulators onto Rocket Secure Host Access. So unlike the capacity priced control plane software elsewhere on the mainframe, the cost here is driven by people, not processors, and the meaningful variables are the seat count, whether the basis is named or concurrent, and whether you sit on a legacy perpetual entitlement or a Rocket subscription.

Directional and pattern level. Confirm the named versus concurrent basis, the active seat count, and the consolidation and subscription terms in your own entitlement before modeling a renewal.

The first driver is the seat count, the headline on any per user product, and the number that quietly grows as desktops are imaged and reimaged over years without the entitlement being reconciled down. The second is version and product sprawl, because an estate that came through the Micro Focus portfolio often runs more than one emulator side by side, each with its own entitlement, so the same user can be counted twice across overlapping lines. The third is the contract model, since the move from perpetual plus maintenance to subscription changes the cash profile and the renewal arithmetic, and the consolidation onto Rocket Secure Host Access is the vendor's preferred path, which is not automatically the cheapest for the buyer. The fourth is deployment topology, where Citrix and virtual desktop estates multiply effective seats in ways an old perpetual license never priced. Session volume is not the cost; seats, sprawl, and the model are.

Emulator exposure is mostly seat counting and version sprawl. Common traps we see at pattern level:

Because Host Access is a per seat product in mid consolidation, the levers are about counting honestly, collapsing sprawl, and the contract model. The five that pay:

Terminal emulation is one of the more switchable mainframe access layers, because it sits on the desktop rather than deep in the host control plane, and competing emulators from IBM, Rocket's own BlueZone, and other vendors all deliver the same core 3270 and 5250 sessions. That makes a credible alternative genuinely real, and it is why displacement campaigns are common in this category. The friction that remains is not the host but the desktop: macros, keyboard maps, scripts, single sign on integration, and the sheer logistics of repackaging and redeploying an emulator to thousands of managed endpoints. The practical approach is to establish seat truth, collapse the sprawl, and choose the contract model first, where most of the saving sits, and to keep a costed move to a competing emulator as real leverage that disciplines the renewal. Where a desktop refresh or a security driven standardization is already happening, the migration cost can ride on that program rather than standing alone.