Skip to content
IRIS NEURALContact
Back to the data centre

CAM-557

Water falling in the aisle and creeping towards the floor grilles

Four in the morning. A thin jet of water is falling from a downpipe running beside the pillar, and there is already a sheet of it spreading over the floor tiles. On one side of the aisle stand the cooling units; on the other, the row of cabinets and the grilles the air comes up through. A technician is standing a few metres away, checking something, and has not seen it yet. IRIS sees the shine of water on the floor, how much the patch has grown in the last few minutes and how far it is from the nearest grille.

Four in the morning. A thin jet of water is falling from a downpipe running beside the pillar, and there is already a sheet of it spreading over the floor tiles. On one side of the aisle stand the cooling units; on the other, the row of cabinets and the grilles the air comes up through. A technician is standing a few metres away, checking something, and has not seen it yet. IRIS sees the shine of water on the floor, how much the patch has grown in the last few minutes and how far it is from the nearest grille.
CAM-557IRIS VisionLive

Turn on IRIS Vision

AI-generated image

CAM-557 · PLANT AISLE · WATER ON FLOOR

What IRIS understands

What it recognises in this scene

  • sheet of water on the floor
  • jet of water falling
  • downpipe beside the pillar
  • cooling unit
  • raised-floor grilles
  • technician checking the room
  • panels at the end of the aisle
  • row of cabinets

What this camera measures

An alert says something is happening. The reading says how much.

  • 4 minWater in sight for
  • 3 m²+2 m²Wet area
  • 1,2 mDistance to the nearest grille
  • 1People in the aisle
  • 03:41:22Video starts at

The figures on this screen are an example from the illustration, not any client's measurement.

The rule

You decide once. IRIS applies it always.

If

water appears on the floor of a plant room and is still there after a minute

Then

  1. alert the duty shift with the camera, the aisle and the time
  2. open the live video of that camera in the control room
  3. save the video from one minute before the water appeared
The instruction that set it up«Tell me if water shows up on the floor of any plant room, even a little, and tell me which aisle.»

What changes

What changes for whoever answers for physical security on site

Water in a data centre makes no noise. It gets found when it reaches somewhere that matters, and by then it is not a puddle: it is a breakdown, an insurance claim and a whole night of people working. With this camera the warning goes out while it is still a small puddle, with the exact aisle and the video of the minute it started falling. Somebody closes a valve and puts a bucket down, and that is all that happens that night. Here is the limit, and it matters: IRIS sees the water that is in plain sight. A drip inside the raised floor, as long as it does not surface, is not visible to anybody through a camera — that is what floor sensors are for. What the camera does is cover the whole aisle, which is far bigger than the spot where a sensor sits.

What people usually ask

Can it mistake a reflection on the floor for water?

It can, and it is better said plainly: a polished floor shines, and at four in the morning it shines more. That is why the rule does not fire on a glint: it waits for the patch to stay and to grow, which is what a reflection does not do. And that is why the alert goes to a person with the video in front of them: checking takes ten seconds. One alert too many at four in the morning is far cheaper than a puddle nobody sees until eight.

And what if the water is under the raised floor?

Then IRIS does not see it, and that has to be said as it is. A camera sees what is in plain sight: while the water does not surface through a joint or a grille, there is nothing to look at. That gap is covered by the raised-floor sensors, which are there for exactly that. What the camera adds is the rest of the problem: as soon as the water reaches the surface, it is seen along the whole aisle and not only at the spot where a sensor sits, and there is video of where it started.

So does it replace the water sensors?

No, and it does not claim to. A sensor warns where it is placed, at night and with nobody around; a camera sees the whole aisle, but only what shows. They are two different things and they get along: the sensor says there is water at its point, and the camera says where it is coming from, how much it has grown, with the video of the moment it started. Whoever has sensors keeps them; whoever does not, at least stops finding out in the morning.

A different case every time

Everything comes in. Only what matters gets through.

This sums up what IRIS does with one alert, from the moment it appears until it is closed. The real screen holds far more; this is the main thread.

  1. Cameras on the map
  2. You draw an area
  3. What passed shows up
IRIS NEURAL
30ES
IRISDirectoGrabacionesGISIncidencias3996SituaciónCasosLPRAutomatizacionesIRIS DATA
Ask a question or make a request…Send
IRIS · Infinity Neural

And the best part

With two screens you watch.With twenty you look without seeing.

It is nobody’s fault: it happens to everyone. IRIS watches all twenty at once and only interrupts when something deserves a person’s attention.

Tell us your problem and we will say whether IRIS understands it, or not yet.

We reply the same working day.