Marlin Data Fabric (MDF)

Automated Network Discovery & Data Extraction Engine

Scope

Marlin Data Fabric (MDF) is an end-to-end data framework designed to seamlessly orchestrate the entire data lifecycle—from ingestion and migration to transformation, quality assurance, and AI-driven intelligence. Built for modern data ecosystems, Marlin Data Fabric provides a unified layer that connects disparate data sources, automates complex workflows, and ensures data is trusted, clean, and ready for advanced analytics and machine learning.

 

 

Default datasets — out of the box

MNI ships with ready-made geographic datasets and dashboards, so the business sees the network the moment it's connected:

Consumer types — connected vs total

A geographic layer of every consumer type — residential homes, businesses, antennas and more — flagged as fully connected (spliced and patched) to the PoP. Powers a dashboard of connected versus total consumers, with a breakdown per region.

Rack & splice-closure utilisation

A geographic layer showing used and available splices and connectors within racks and splice closures.

Duct utilisation

A geographic layer of occupied versus available microducts along the route — a microduct is “occupied” once a cable is installed in it.

Cable utilisation — used vs available

A geographic layer of occupied versus available fibres in each cable — a fibre is “occupied” when it’s part of an end-to-end connection between a consumer and a PoP — plus a layer aggregating utilisation per cable type and route (distribution versus backbone).

How good is your data?

Under the hood: AI/ML and computer vision read unstructured legacy — spreadsheets, PDFs, Visio, diagrams — alongside geospatial tooling, always with a human in the loop. Start with a Data Assessment, or see how we use AI.

Ask for a Data AssessmentHow we use AI?

Book a demo

A focused demo, mapped to your network and your processes — run by someone who has migrated networks like yours. Or start with the question that changes everything: how good is your data, really?