JSI · Physical Security Engineering · Project EGA

How Many Cameras Fit in a Trailer

Four hours of continuous measurement on the trailer gateway, running seven cameras through EGA and two through GeoVision at the same time. What the hardware can carry, what it cannot, and where the energy actually goes.

Machine
Intel N150 · 4 cores · 16 GB
Run
4 h continuous
Samples
191
Date
2026-08-22

The short answer

Nine cameras through EGA is comfortable. Nine cameras through the VMS is not possible on this hardware.

The gateway is the cheap half. Each D1 camera costs EGA about 0.21 of a processor core; the same camera costs the VMS about 0.36 — roughly seventy percent more. The limit on a trailer is not how many streams we can annotate, it is how many the video management system can record and watch.

5.2% Per camera, EGA

Of the whole machine, for one D1 stream: decode, draw, encode, publish.

9.0% Per camera, GeoVision

The same camera costs the VMS nearly twice what the gateway costs.

74% Average total load

Today's 7 + 2 configuration, with peaks reaching 97%.

0.0 GB Memory drift in 4 h

8.35 GB at the start, 8.29 at the end. Nothing leaks.

What was measured

The trailer gateway ran its normal workload for four uninterrupted hours: seven cameras pulled, annotated and republished by EGA, and two of those same cameras recorded and monitored by GeoVision on the same machine. Samples every ten seconds, 191 of them.

MeasureMeanMedian95th pctPeak
Total CPU74.0%74.0%89.5%97.0%
EGA (7 cameras)45.7%45.6%48.7%60.1%
GeoVision (2 cameras)17.9%17.1%26.9%32.5%
Everything else10.4%
Clock speed1718 MHz1718 MHz1794 MHz1846 MHz
Memory in use8.2 GB8.2 GB8.3 GB8.9 GB

Seven cameras stayed connected for the entire run. No stream dropped, no process restarted, and memory finished six hundredths of a gigabyte below where it started. The software is stable; the question is purely one of capacity.

The finding that matters

We expected the gateway to be the expensive part. It is not.

Cost of ONE camera, in processor cores EGA 0.21 cores GeoVision 0.36 cores — 70% more Where the four cores go at nine cameras each EGA ×9 — 1.9 cores GeoVision ×9 — 3.2 cores ← the machine runs out here (4 cores)
Together the two would need 5.1 cores on a machine that has 4. The gateway is not what runs out first.

This is worth stating plainly because it inverts the intuition. Annotating a stream — decoding it, drawing on it, re-encoding it, republishing it — turns out to be cheaper than what the video management system already does with the same stream today.

What fits

Every figure below adds the fixed 10.4% the operating system and its services consume regardless of our software.

ConfigurationTotal loadVerdict
EGA 7 + VMS 2  (today)65%Works, thin margin
EGA 9 + VMS 275%Works, thin margin
EGA 12 + VMS 291%Do not
EGA 9 + VMS 493%Do not
EGA 6 + VMS 696%Do not
EGA 9 + VMS 9138%Impossible

The measured average of 74% is close enough to saturation that peaks already touch 97%. A system with no headroom has no room for a camera reconnecting, a Windows update, or a detection burst when three people cross at once.

Recommended limit

Nine cameras through EGA, two through the VMS. That lands near 75% average, which is workable but should be treated as the ceiling rather than the target. Anything beyond it needs one of the changes in the next section.

Where the energy goes

On a solar and battery trailer the useful question is not whether the processor copes but what it draws. One measurement speaks directly to that.

The chip never rests

The N150's base clock is 800 MHz. Across four hours it never once dropped below 1568 MHz, averaging 1718 — more than twice base, continuously. The processor spent the entire run in its boost range and never returned to a low-power state.

Power on this class of processor rises faster than clock speed, so a chip held at twice base frequency all day costs disproportionately more than the utilisation figure suggests. This matters more than the raw percentages: reducing load enough that the processor can step down to a lower state saves more energy than the utilisation drop alone would imply.

That reframes the target. The goal is not to fill the machine to 90% because it fits — it is to keep total load low enough that the processor is allowed to idle down between bursts.

Practical levers, in order of value

  1. Move cameras off the VMS, not off EGA. Each camera removed from GeoVision frees 70% more capacity than removing one from the gateway. The cameras already record to their own SD cards in full quality, which can be retrieved on demand — so VMS recording may be redundant for some of them.
  2. Resolution, at the camera and not in software. Reducing a stream costs nothing if the camera sends it smaller; doing it in software saves only half the work and adds a scaling step. Note that VGA is only 9% smaller than D1 and not worth the effort — the meaningful step down is CIF, at 70% fewer pixels, which does visibly degrade what an operator can judge.
  3. The pan-tilt-zoom camera costs as much as three others. It refuses D1 on its third stream and runs at 720p, which is 2.7× the pixels. It alone accounts for 0.57 cores of EGA's 1.83.
  4. Windows Defender. It was among the largest single consumers during the run. Excluding the gateway's own folders from real-time scanning is free and safe; disabling it entirely is not advisable on a machine that currently holds a public address with its firewall switched off.

What this does not tell us

Four honest limits on the numbers above.

Recommended next steps

  1. Measure the VMS with more cameras. Add channels one at a time and watch the curve. This is the number the whole capacity question rests on, and it is currently an estimate.
  2. Pair the load curve with the power readings the team already takes, at 0, 3, 6 and 9 cameras. That converts this report from percentages into watts per camera, which is what a battery budget actually needs.
  3. Decide whether the VMS needs to record every camera, given each one already records locally in higher quality than anything re-encoded. This is the largest single lever available and it is a policy decision, not a technical one.
  4. Repeat the run overnight, during real monitoring hours, with boxes actually being drawn. Today's measurement covers the quiet case.