Repository navigation
Replies: 1 comment 1 reply
|
Agent memory typically sits between 5 and 25 MB depending on available memory and what you're monitoring. Anecdotally, I've never seen CPU average more than 0.05%. It can run easily on low powered devices like routers / SBCs. As an example, I have a VPS with 6 shared Ryzen 9 9950X cores that runs/monitors 127 containers. The agent sits between 15-20 MB ram and uses 0.01% CPU. My personal hub runs on an old Hetzner VPS with 3 shared AMD EPYC Rome cores, has ~25 agents connected at any given time, and uses 90 MB memory and 0.06% CPU. Leading solutions I tried before making Beszel are the Prometheus/Grafana/Cadvisor stack, Netdata, and Checkmk, which were all significantly heavier. Please note that I don't call them "heavyweight," as this is completely relative, and all the tools have different capabilities. I only say that Beszel is lighter, since that was my main reason for making it. This was in 2023/2024, so please feel free to make your own comparisons and report back. |
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
The https://beszel.dev/guide/what-is-beszel page says leading solution are heavyweight, but no prooflinks. Does anybody know what are those and how bigger and more resource-intensive they are?
EDIT: Also, what is the actual
beszelagent memory and CPU footprint?All reactions