VECTORIO“Near-real-time RIC loops need answers before the batch pipeline wakes up”
Real-time RAN analytics at the edge.
Real-time features, every sample kept
- xApps need answers in tens of milliseconds, my pipeline works in batches.
- Shipping all RAN telemetry to a central cloud is too slow and too costly.
- Multi-vendor Open RAN means multi-vendor data formats and nobody owns the data layer.
- Real-time features, every sample kept
- Less traffic on the bottleneck links, and no lost events when the network splits
- Controlled processing across trust boundaries
From today to the first result.
batch pipelines on a central cloud
The near-real-time RIC acts on sub-100 ms intervals, while the pipeline delivers its answers in batches.
Features are computed at the edge while the stream runs, with checkpoints and replay. The engine keeps a 10,000-node workflow in real time (measured on audio at 48 kHz).
Run one telemetry feature next to the batch pipeline on a test cell.
one KPIWindows over a live event stream. Step by step.
Events are grouped into windows as they arrive and tagged, ready for real-time dashboards and alerts.
- Event DataRandomDataProvider
- Tumbling WindowWindow Functions
- Tag Window NumberCompute Fields
- Windowed Events CSVCSV Publisher
event-gen → Generators.Random
tumbling-window → ETL.Window
window-tagger → ETL.Compute
windowed-csv → CSV.Publisher- Event DataGenerates seeded, repeatable test rows.
- Tumbling WindowComputes window functions and row-count windows.
- Tag Window NumberComputes new columns from expressions.
- Windowed Events CSVWrites rows to a CSV file.
A real workflow from our end-to-end test suite; the designer view is illustrated.
What you get. From the first pilot.
Real-time features, every sample kept
Less traffic on the bottleneck links, and no lost events when the network splits
Controlled processing across trust boundaries
Why your tools fall short.
Kafka/Flink/Spark chains carry latency and footprint unsuited to edge RIC placement.
Vendor-proprietary analytics lock operators to the RAN vendor.
Cloud analytics add backhaul cost and sovereignty exposure.
- SNS JU (Smart Networks and Services Joint Undertaking) 6G research programme
- O-RAN ALLIANCE specifications (near-RT and non-RT RIC)
- EU 5G/6G security and sovereignty toolbox
Your pilot. Three steps.
Your data shape
Tell us what data you have, where it lives and what the result must be.
A scoped pilot
On your data and your hardware, with the success measure written down first.
A clear decision
You see the result against that measure — and keep every workflow and record.
Related roles in Sovereign & edge.
Bring us your hardest data problem. Let's solve it together.
A short pilot on your data and your hardware. We agree on what success looks like before we start.