Testing like you mean it

CollatrEdge has 800+ tests. More lines of test code than production code.

That is deliberate.

The E2E suite runs a full pipeline: real config parsing, real plugin instantiation, real data flow from input to output. The sustained operation tests run …

Read more

Processing data at the edge

Raw sensor data is noisy. A temperature probe that reads 1,847 times in an hour does not need 1,847 rows in a database. It needs a min, max, mean, count, and standard deviation. One row.

CollatrEdge runs processors and aggregators locally. Before the …

Read more

Data does not leave the building (unless you say so)

Every CollatrEdge installation has a local SQLite store. Always on. Not optional.

This is the thing that most IIoT platforms get wrong. They treat local storage as a fallback. Something that kicks in when the cloud connection drops. An afterthought. …

Read more

Reading from factory equipment

Modbus TCP has been around since 1979. It is still everywhere. PLCs, VFDs, power meters, temperature controllers.

The CollatrEdge Modbus plugin reads coils, discrete inputs, input registers, and holding registers. It handles four byte orders (ABCD, …

Read more

The review loop that catches bugs before they ship

Every phase of CollatrEdge goes through the same process. Build. Test. Internal code review. Fix. Independent code review by a separate reviewer who has never seen the code. Fix again.

Phase 1 found 13 issues in the first review. Three were red: a …

Read more

The ticker problem

Factories run on clocks. Shift starts at 06:00. Readings every 30 seconds. Reports aligned to the hour.

CollatrEdge uses two clocks. A monotonic clock for intervals (immune to NTP adjustments) and a wall clock for alignment (so your 30-second …

Read more