Update Proposal for Storage Node Operators

Storj would get more free space when the temp folder is emptied on every start and satellites cleaning up after themselves when they DQ a node. Also moving trash data instead of copying might help in the big picture.

What I mean, that in the grand scheme of things test data is not important, it should not be. How big is the test dataset, 6.5PB on nodes? That’s about $10k per month for storage. If this amount of money is important enough to risk losing oldest, and most experienced SNOs, then I would not be very optimistic about the future of Storj. But if this is the message you don’t mind sending, then okay.

As for the pricing model, it does not really matter, provided the resulting income is acceptable and predictable.

p.s. For me personally it would be absolutely great if you drop test data right now as I could then solve my migration problem without moving too much data around and probably even GE instantly to recover a significant amount of money.

I think the cut in payouts is going to remove more nodes than removing the test data. I doubt anybody leaves because they gained some space on their drives, even if it is a good portion. As BrightSilence points out, the data growth is strong right now. If we cleared that data out, most of you that have old data would get it back after some time as incoming data is moving at a good clip. And assuming this is data that is being pulled on, you would earn more from it than the stale data on your drives.

6 Likes

Let me say that this whole situation and how we all would like to understand the details of it is very strange for me… 90-95% of the whole discussion is guessing about the current situation of Storj.
Yes there are excellent ideas, but otherwise we are “fighting with the fog”, so we don’t have any real info about the details. We already have 695 replies in this topic and noone ever wrote down that how much of the currently stored data is real customer data, how much of it is coming from the free tier and how much is test data.
We are requested to comment a business plan here, but with very limited information. We try to run calculations based on guessings. This is non-sense. In various forum topics we are runing the same circles again and again…

Decission making based on open forum feedbacks while the company is dealing with millions of USD in the balance sheet… I’m sorry, but this is everything but professional.

3 Likes

i for one will keep adding storage :smiley:
free space is a bit relative, and test data a buffer…
i think we should just disregard the test data as it will just keep becoming a smaller and smaller % because it is mainly a buffer for storj to onboard large customers…

1 Like

Why is free tier not real customer data? I have data stored but only a small amount, does that mean I’m not a customer?

https://storjstats.info/d/storj/storj-network-statistics?orgId=1

EU1 + US1 + AP1 = customers
Saltlake + US2 + Europe-North = test data

This is mentioned above.

2 Likes

Careful what you wish for… What might be considered professional is sharing nothing and making decisions behind closed doors. In which case… imagine how little I care about it looking professional. I’m glad they are showing a proposal first and based on the twitter space, it has directed their decisions. And while we don’t know everything, we know plenty to have had very informed discussions in this topic. We now basically know almost everything related to COGS thanks to some helpful replies and even if we don’t know how much free tier use there is, we can just assume that part will eventually be solved and ignore it for this conversation.

For my biggest and oldest node it’s about 2/3rds. But I also have a smaller node that filled up before there was any customer traffic, for which it’s like 90%. I’m not saying it should take a year, just saying stretch it out over 2-3 months so a good chunk of it is replaced at the same time.

I’m honestly baffled to see no deletes at all this month while ingress is approaching 2TB per node. It should have definitely started in force this month.

4 Likes

Nope. We were not asked to comment a business plan. We were informed of a radical cut to our pay and asked how we feel about it, so they have an idea about how disruptive such a cut might be.
In the end, the payout will be the maximum proposed, or even a bit higher. You know, “Lower the expectations and then offer a treat”.

Business is not about feeling. They asked about our running costs and what will be affordable payout to run nodes. So what is your expenses on running nodes?
Different countries have different electricity, internet costs.

3 Likes

Just to add my 5 cents…
A few months ago I started to operate a SN with 3.5 TB. My setup is a recent NUC with a Thunderbolt 4-bay HDD case. I have TrueNAS Scale installed and use it mainly as my NAS. I installed two old 4 TB HDD (ZFS mirrored) for sharing with StorJ. The additional power consumption accounts to 10-15 Watts (or 11 kWh per month costing me ~3 €)
I have 1 G Fiber symmetric without traffic restrictions, so no cost from the traffic side.

I would prefer to have the storage price stay at $1.5 to cover for the electricity bill for sure. Egress would than be the profit. I fine with the upper limit of the proposed payout ($5). Audit and Repair could also be $0. Actually I would also be happy to get a storage quota in return for my on needs (mainly Backup) instead of “money”. A conversion rate of 1:3-5 would be ok I guess.

I believe that StorJ should be about sharing excess resources from the community for the community. Not “professional” large scale SNs. The day I get the impression that as SNO I’m just fueling company profits (of StorJ or it’s customers), I’m out. No matter the price.

3 Likes

I strongly second this. Having decent payment for storage means the operating costs are covered, maybe even allowing for some expansion or recovery after catastrophe. And then egress is just a bonus.

As I stated previously, I would be fine with repair being free, cause healthy network means my nodes are going to live and prosper too. Though some have issues with this being free - probably those who run in datacenters with paid traffic :upside_down_face:

And I have the same approach overall as you. Storj Labs is not a non-profit organization, it’s not a Folding@home type of thing, so our payment should come before company’s. If that means the network doesn’t survive, I would be quite sad, but I understand that not every business is profitable in the real world scenario.

3 Likes

I beg to differ. Business is all about feelings…
My running cost used to be zero, since I was running a node in space I had available in my NAS. But the unfortunate BTRFS choice meant that once the node got big enough, I couldn’t use my NAS properly. I then decided to buy a 14TB disk and move the node there. This was my first “storj expense”. It’s ok, the money I had made with the node was just enough to pay for the disk. Then I decided to get rid of an old NAS I had running. I ended up with 5 4TB disks. I decided to put these disks to use and deployed 2 nodes in different locations (family). So, the extra disk in my NAS is running at 9W. That amounts to ~1.5€/month. I’m still good on that one and will still return a profit with the proposed cuts (because I would run the NAS even without storj). My family is not asking me for electricity costs, but I do have a conscience. One of the nodes is not making enough to cover electricity, and with 2 4TB disks, hardly ever will (with proposed payouts!!!). The other node just has no chance. It will never cover electricity costs using 4TB disks.

PS1 - Notice that it is quite ok to do the math on electricity costs when you want to estimate profits. But when you are forced to do the math on electricity costs just to conclude if you can have a profit or if you are loosing money, it’s kind of ridiculous…

PS2 - I don’t think that storj can survive like this. It’s just a matter of time. When your disks blow up, you won’t be able to buy new ones. Also, I acknowledge that storj wouldn’t survive long term with the present payouts. Something had to be done.

PS3 - Presently, storj can afford loosing a bunch of nodes. Let’s see what the future brings…

1 Like

fair point… though I really hope it isn’t all free tier… Would be really helpful if this could be quantified. I’d love to know the impact of that.

In my opinion current free tier is too generous. And let me explain. I have uploaded to storj about 30GBs of tv series that I download at night an episode (apx 300-400ΜΒ) before bed on my phone to watch them. I already transfered to Storj 15 dolars in Storj coin in January.
Since January I have never exceeded the 1.65 coupon applied to my account. Not with traffic, not with storage. And 15 dollars just sits on my acount.
And my usage is only one example.
Storj needs 5GB free space for free tier and 5GB for bandwidth. And that’s it. I50GB storage is too much and of course attrackts abusive behaviour.
In a side note, I have 10TB files in cold archive in Alibabacloud. And i tried many times moving them to storj, but the 40 euros for 10TB are too much. My proposal

Free tier:
5GB storage and 5GB bandwith.

Storj for customers
3 euros per month storage (mega.nz launches soon Mega s4 for 3 euros/TB if I remember correctly.
0.01 euro /GB download bandwith.

Node operators
10 euros per TB Bandwidth
1 euro per TB storage
0 euros audit
7 euros repair.
anything less than that poses a risk to the network. I am not telling that its going to be profitable for Storj, but try these numbers first before going for the prices you mentioned at the first post for Storj Node operators.

Sorry for my english.

3 Likes

No need, I already agree. I’m in a similar boat. I have about 100GB of personal backups on Storj and $50 sitting in the account that is no longer being used due to the free tier. I was all ready to pay for it and then it suddenly wasn’t necessary anymore. The service is good, it deserves to be paid for.

Based on the responses on the twitter space they recently did, I expect they will do exactly this. Lower it gradually to see the SNO reaction. They’d be dead in the water without node operators, so they know they need to be careful.

In the third quarter of :point_right:2022 :point_left:, SNO received 0.5 million tokens out of a total expenditure of 18.5 million. Thus, it is really not 5%, but even less - 2.7%.

2 Likes

So? It’s a start up. The terms of being paid as a SNO were and are known. SNO’s have earned more than revenue. You seem to be missing that fact. Not only that, but Storj Labs has provided synthetic data to spend even more money on SNO’s.

What the company pays for people and resources is expensive. This isn’t three guys in a basement.

And the numbers are still outdated. Move on.

4 Likes

There is no more recent public data, otherwise I would refer to them.

1 Like

I kinda object to this. Storj has given no money to SNOs.
They have used a token to transfer payments or something of equal value to money

1 Like

Symantics. Tokens fixed to USD in value at the time of payout. For the purpose of conversation brevity we aren’t going to break down the way the transaction model works. Everyone here is aware of how they are getting paid.

2 Likes

But they are made by the company itself for free. So technically the company doesn’t spend anything.

3 Likes