The operational problem
The flow is designed, but executing it takes more than opening a browser: reading the file, reaching the data server and having the right dependencies available. A valid laptop path may not exist in the environment that actually processes the load.
Without checking that difference, infrastructure failures can be mistaken for Mapping errors. You need to know where the process runs and whether that runtime meets its execution requirements.
How Dova helps
Dova uses workers to execute compatible workloads. The inventory shows reported state, capabilities and resources; runtime selection checks eligibility, recent connectivity and available capacity, among other conditions.
A Session can use the tenant’s default selection or an explicit worker when configuration and permissions allow. This separation helps prepare the execution environment and investigate access problems without unnecessarily changing transformation logic.
How it works
Identify the pipeline requirements
List components, connectors, paths and external systems. Define which files the execution environment must read or write.
Review the registered worker
Open Workers and check state, recent activity and capabilities. An inventory entry does not prove the worker is enabled or able to execute your workload.
Check access and dependencies
Verify process files and network access from the worker’s environment. Prepare connector dependencies and authorized credentials using the intended mechanisms.
Configure the Session
Keep the default runtime or choose a compatible override when allowed. Review profile, bindings and versions before publishing the context.
Execute a minimal test
Run a laboratory Workflow and inspect Monitor. Confirm where the output was written and document the runtime before increasing the workload.
What you need
- A registered, enabled worker eligible for the Session’s context and the Mapping’s capabilities.
- Network, file and credential access to authorized resources, plus the dependencies required by the connectors.
- Permission to configure runtime and execute; synthetic data, isolated destinations and an expected result.
Limits to consider
- Online indicates agent activity under its contract; it does not guarantee capacity, access to every system or an execution deadline.
- Registering or selecting a worker does not copy files from your laptop or automatically install every connector dependency.
- Reported slots and resources help evaluate capacity. This feature does not promise autoscaling, immediate execution or universal resumption of interrupted workloads.
Demonstration example
Illustrative example using synthetic data. This is not a customer result or a product screenshot.
Demonstration: process a CSV on the selected runtime
Prepare a synthetic CSV and an output directory accessible from the laboratory worker. Bind both to a normalization Session and execute the Workflow. Check both the attempt in Monitor and the file generated in that environment; this does not yet validate operational performance.
Accessible synthetic CSV
↓
Session + compatible worker
↓
Test file + inspection in Monitor