Paid Storage Growth: Two views?

[Roxor]

Todays Town Hall shows more than double the paid-used-space as a year ago. But the Satellite stats show we’re effectively the same place for “stored customer data” looking back that same year. What gives?

Maybe deletion of all the old forever-free data has been offset almost exactly by real customer uploads? (If that’s true, then Storj counts what-they-pay-SNOs-for as “stored customer data”, because they compensate us for it: even if they didn’t get paid themselves?)

Just interesting to see. Maybe we’re on the cusp of seeing some of that 11%-compound-monthly-growth in our nodes too?

[naxbc]

Well, I can say this…2 out of 5 nodes (10TB used space) are Gracefull Exited. It´s not even paying in electrical consumption (HP N36L and N40L).
Might as well Exit, sell the hardware and buy BTC.
RIP, SNO´s

[arrogantrabbit]

@naxbc

It´s not even paying

@naxbc

Might as well Exit, sell the hardware and buy BTC.

How about we stop repeating this dumb shit?

For those under the rock: Don’t keep hardware online just to host node. It’s literally the opposite of what this project is attempting to accomplish.

I strongly recommend reading some docs before embarking on any project.

[littleskunk]

@naxbc

It´s not even paying in electrical consumption (HP N36L and N40L).

I had the same problem and decided to downside my node to a Pi5. That reduced my electricity costs a lot and the payout keeps increasing. Not as fast as I was hoping for but enough to make it worth it.

[naxbc]

Yeah, tried it, even have a RPI5 I bought for that effect but the overall return is around 8$ before electric cost, even with a Pi.
We must take under account that a PI even being the 5, just can’t handle the charge and uploads take a beating

[jammerdan]

@Roxor

Maybe deletion of all the old forever-free data has been offset almost exactly by real customer uploads?

@Bryanm

As previously mentioned when we announced the discontinuation of the free tier , We have also already been removing old data from our production satellites composed mostly of old abandoned accounts as well as free tier accounts that were closed by users following the conversion to free trial on our production satellites. We plan to increase the rate at which we are doing this in order to transfer all non-paid data to the SLC satellite and clean up the production satellites. As a consequence of this, nodes will see an increase in the amount of garbage they are storing in the short term.

Maybe this is what we see in the Grafana chart like from April to June. Unfortunately the graph has so many glitches and there were so many errors in displaying correct numbers that it is hard to tell if the values are even correct. Some of the lower marks is showing as 15PB stored customer data I doubt we have been this low. But maybe the 23PB are correct. This would mean that we gained around 2 PB since then. However we really still don’t know how much and at what pace old data has been deleted or may still be getting deleted. I am also wondering where is the 10PB customer in that chart?

[Alexey]

@Roxor

Maybe deletion of all the old forever-free data has been offset almost exactly by real customer uploads? (If that’s true, then Storj counts what-they-pay-SNOs-for as “stored customer data”, because they compensate us for it: even if they didn’t get paid themselves?)

Yes, we slowly removes the frozen accounts with the growing of the paid customers data usage.

@jammerdan

I am also wondering where is the 10PB customer in that chart?

it should be on US1

[jammerdan]

@Alexey

it should be on US1

What I mean is: Shouldn’t a customer with 10PB somehow be visible in that used space from the graph: Grafana

I cannot spot it.

[Alexey]

I think the satellites filter there doesn’t work. I think you can try to add an own panel with hardcoded filter for US1 only. Or fix the picker

[jammerdan]

@Alexey

I think the satellites filter there doesn’t work.

There is no specific filter applied.

According to this statement:

@littleskunk

This metric includes the storj select nodes we are adding to the network.

I understand it that the Select data is included in that graph and I would expect that +10PB of data would somehow visible.

[Alexey]

It’s hard, because we have many customers, 10PB only a small part of the usage.

By the way, the fix for a satellites picker is pretty simple
Query B

max(storj_stats_storage_free_capacity_estimate_bytes{satellite=~"$satellite"})

Query A

sum(storj_stats_storage_remote_bytes{satellite=~"$satellite"})

[EasyRhino]

does storjstats include storj select? I thought the 10PB customer went to storj select.

[jammerdan]

@EasyRhino

does storjstats include storj select?

Updates on Test Data

This metric includes the storj select nodes we are adding to the network. I hope at some point the metrics will differenciate between public and storj select network. Right now they don’t and to make things worse there are even storj select nodes on all satellites. The best reading you might get from the EU satellite since there the number of storj select nodes is still low.

[daki82]

@Roxor

Maybe deletion of all the old forever-free data has been offset almost exactly by real customer uploads?

So after some months of no growth, my fast node finaly catched some 300GB ish data more on the disk.
(W24TB are waiting offline, maybe i get an second internet line as “backup-line ”)