If you feeded the yearly graph, then you gave it the bandwidth of the load test as well. Did you specify that it was a test period, where no real data was uploaded at high speeds?
And for the info, that the network is growing, you don’t need an ai for. The statistics alone are saying that.
Yes, i told AI to not include test period. Gemini deep research is very powerful for things like this
Been a day now…egress continues.
I dunno, there’s been another slaughter of TTL data today, but really I’m enjoying watching the network having a workout - that’s what it’s for.
2 cents,
Julio
me too. Nothing special in the last weeks.
Is it only a small set of pieces that get downloaded very often? Maybe your node was lucky to get one of the famous pieces then 
1 Like
It’s weird, one of my testing machines, at one of my remote sits, holds 25 nodes. All egress on happened on a single one of them. Pretty crazy to see 6GB ingress but 150GB egress in a single day
2 Likes
Only one out of 80 nodes showing that egress rush. Seems like someone is downloading the same data again and again for whatever reason.
2 Likes
for me High egress was only on 10% of nodes.
No change in either direction here.
Excellent feedback…I have more to share responding to Vadim in a minute here.
2 cents,
Julio
Moreover my note is that there was one node winner (egress) for every /24 ip these systems have unique from one another.
Hmmm,
2 cents,
Julio
Ottetal - mind sending me a message?
I tried sending you one, but an error happend saying you dont receive messages at the moment.
Cheers
I’m very late to this thread, but just looked at http://th3van.dk/ today (2025-10-21), and saw these metrics: in the top section:
Total used space - reported by SN ------- : 531.44 TB
Total amount of trash - reported by SN – : 21.56 TB
…and in the HDD Overview section below:
(cumulative) Used hdd space: 730.345.678.077.952 (27.5811 %)
Is the ±180TB difference between filesystem-used-space and node+trash-used-space due to the switch to hashstore? Like hashstore data-logs are all holding a bit of uncompacted data… that perhaps isn’t even showing up in the node-trash number? That’s too large of a difference for filesytem/metadata-overhead or base2-vs-base-10-reporting to explain.
I guess it’s not important to understand… until the disks are 100% full. Just interesting…
1 Like
after i updated to latest version my gone much better almost 1 to 1
2 Likes
I found out that the Total used space - reported by SN was using 1024 instead of 1000, in the script I’m using to fetch the data from my nodes.
Thank you for letting me know about it 
If you like, I have a very simple page showing how much data is used by each node, with a total at the bottom of the site :
Node used hdd space overview
Th3Van.dk
9 Likes