The Carve Out Approach to Oracle Java.
Keep Oracle Java only where a workload truly needs it and move everything else to a free distribution. The carve out is the most common middle path between a full renewal and a full exit, and usually the cheapest.
The carve out approach keeps Oracle Java only where a workload truly needs it and moves everything else to a free distribution. It is the most common middle path between a full renewal and a full exit, and usually the cheapest.
The idea behind a carve out
Most enterprises do not need to choose between renewing the whole Oracle Java subscription and abandoning Oracle Java entirely. There is a middle path, and for the majority of estates it is the right one. The carve out approach isolates the small set of workloads that genuinely depend on an Oracle distribution, keeps those on a tightly scoped Oracle agreement, and moves everything else to a free OpenJDK build. The result is a licensed footprint that reflects real need rather than the accident of how Java spread across the estate.
The metric is what makes the carve out so valuable. Since January 2023 Oracle has priced Java SE on the Universal Subscription at 5.25 to 15.00 dollars per employee per month, counting every full time and part time employee, every contractor, and every temporary worker. The subscription does not price the Java you run, it prices your headcount. So the goal of a carve out is not to count machines, it is to reach a position where Oracle Java is so contained that the only defensible licensed population is a fraction of the workforce, or where the residual can be handled outside the per employee model entirely.
What carves out cleanly and what does not
| Workload type | Carve out verdict |
|---|---|
| General business apps on standard Java | Move to free OpenJDK |
| In house tools your team controls | Move to free OpenJDK |
| Vendor app that mandates an Oracle build | Keep, or press the vendor |
| Workload using an Oracle only feature | Keep and isolate |
Indicative. The honest finding in most estates is that very little truly requires an Oracle distribution once each dependency is tested rather than assumed.
Why the carve out usually wins
A full renewal pays the per employee price for every workload, including the many that could run free. A full exit removes Oracle Java even from the few workloads that genuinely benefit from it, which can mean fighting a vendor or rebuilding something that worked. The carve out avoids both extremes. It pays Oracle only where there is real value in doing so and stops paying everywhere else. For most enterprises that is the lowest total cost and the lowest disruption, which is why it is the path we recommend most often. It pairs naturally with partial migration as an exit strategy, which is the mechanism that delivers the carve out in practice.
Running a carve out
- Map dependencies. Test which workloads actually require an Oracle build rather than assuming they do.
- Draw the line. Define the contained set that stays on Oracle and the larger set that leaves.
- Migrate the majority. Move everything outside the line to a free OpenJDK distribution.
- Scope the residual. Negotiate the smallest possible Oracle agreement for what remains.
- Lock it with governance. Stop new Oracle builds from leaking back across the line.
A carve out only holds if new Oracle builds do not drift back into the migrated estate. Pair it with a credible plan, as in building a credible Java exit strategy.
The number a carve out produces
When a carve out is done well, the Oracle conversation stops being about your whole headcount and starts being about a contained, defensible slice. That shift is where the savings live. A buyer who arrives at renewal with most of the estate already on a free distribution negotiates the residual from a position of strength, and across our work that produces an average reduction of 68 percent versus Oracle's opening number. The carve out is not a compromise that splits the difference. Done properly, it captures almost all the savings of a full exit with almost none of the disruption.
This is core buyer side work, and it is what we do. We sit between you and Oracle, we never take vendor money, and we draw the carve out line where it serves you rather than Oracle. A Fixed Fee starts from $18,000, agreed up front. Or choose Gainshare, a share of verified savings or avoided exposure, with zero retainer and no risk to you. We have defended more than $120M in Java exposure and over 300 Java audits, with more than 20 years of combined experience on the buyer side.
Where to go next
The carve out is the practical heart of most exit strategies. Ground it in our Oracle Java licensing guide for 2026, then read building a credible Java exit strategy for the wider plan it fits inside. To scope your own carve out line, download the guide and bring your estate to a Strategy Call.
Download the guide.
Get the buyer side OpenJDK migration guide for the full playbook on shrinking your Oracle Java footprint, then bring your questions to a Strategy Call.
Download guide