Capacity is used, not stranded
On-prem clusters, reserved cloud and burst capacity are visible to one scheduler, so idle capacity stops being invisible.
Compute runs every study, twin and test a loop needs on infrastructure you control: cloud, private VPC, your own clusters or air-gapped enclaves. Placement is decided by policy, licences and cost before a job is scheduled, and your existing solvers come with you.
Elidyne does not ask you to migrate. Placement, entitlement and data residency are resolved per job, across the environments your programme is allowed to use.
Before a job is submitted, Compute resolves where it is permitted to run. Data classification, licence entitlement, hardware, cost and current availability all narrow the target before anything is scheduled.
| Target | Control | Eligible |
|---|---|---|
| Managed private / VPC | Joint | Yes |
| On-prem HPC cluster | Customer | Selected |
| Partner GPU pool | Partner | Licence gap |
| Air-gapped enclave | Customer | Yes · offline |
Commercial, open-source, in-house and Elidyne-native solvers run side by side. Bring your own containers, your scheduler integration and the decades-old code nobody is going to rewrite.
The same governed workflows run inside restricted and air-gapped environments. Models, policy and evidence stay within the boundary, and updates move through a controlled, reviewable channel.
Cloud, private VPC, customer on-prem and air-gapped or classified. Every loop, and the control pane around it, behaves the same in each; only the boundary changes.
On-prem clusters, reserved cloud and burst capacity are visible to one scheduler, so idle capacity stops being invisible.
A job that may not run somewhere never lands there. The rule is applied at placement, not discovered in an audit.
Connected, private and air-gapped deployments run identical workflows, so a restricted programme is not a second-class one.
Sovereignty is not a deployment option we bolted on. It is the assumption we designed against.
Six products, one control pane. Compute mainly carries Twin and Test; the rest of the loop is already connected when you need it.
Estimates and reserves compute for each approved plan, and reports spend as runs go.
Open Workbench →02 · Work in ElidyneRuns twin re-runs and forecasts close to the live data.
Open Operations →03 · Work in ElidyneRegenerates figures and tables from stored results.
Open Reports →04 · Under the hoodAnswers Elidyne’s placement and cost questions before a plan is shown to you.
Open Agents →Sovereign compute and solvers, placed by policy
You are hereApplies data classification and residency before any job is placed.
Open Governance →Measure → Twin → Test → Limits → Act → Trace. Every product serves the same six steps, in every loop.
Tell us what your estate really looks like: cloud, on-prem, partner capacity, licences and classification boundaries. That is the version we design against.