If it is eliminated, would it be possible to set it so that a portion of my earnings as a SNO is redirected to pay for it? I do not plan on using the account for anything other than an occasional test or a file transfer, so if it was not free, I would have to pay something like $0.1/month or whatever. This would be annoying if I had to pay transaction fees on top of that.
Something like being able to specify a user name in the node configuration, so when the satellite calculates payment, it subtracts what I owe from my earnings and pays me the remainder.
In february, my node stored 23.16TBm of data and egress was 684GB (GET), 391GB (GET_REPAIR) and 54.91MB GET_AUDIT (I’ll ignore audits in my calculations).
I got $52.34 for it. (or about $2.25 per TB stored)
With the proposed payouts I would get $28.5 (best case) or $19 (worst case). That may be barely enough for me, people with lower capacity (or newer) nodes would get almost nothing. Worst case is $0.82 per TB stored, if my node had less than 10TB of data it would not be worth the effort to run it. Currently, I pay 0.28EUR/kWh for electricity. With $19/month payout, my 23TB node could use no more than ~95W to just break even for electricity.
Assuming a hard drive uses 6W, using drives smaller than 1.5-2TB would not pay even for the electricity used by the drive itself, much less the CPU.
If egress is doubled that would be better, but egress depends on customer demand, so yeah…
If the payouts were like this from the start, I would have never started my node.
So much for decentralization and zero-knowledge encryption…
Why do the edge services not cost more than using uplink?
Oh yeah, “you can spend time to set up the node, make it realiable, look after it and we will pay you $1/month which will be paid out as $24 every two years”.
So the network is doomed to die. When a hd breaks no one will buy it back because it will be uneconomical. If these are earnings per Terabyte, even if customers grow, it will still be uneconomical. Perhaps Storj only wants to keep the big players now, who can scale more easily, and who in any case have more possibilities. So much for decentralization.
And we’re not even calculating fees…
I see this proposal does care about future scale. Do you plan to release any proposal how operators will be able to scale?
This is the point for me. I will invest money and time to get payout. More resources I invest, more scale I have, more I earn. Will storj have a cap on the operator size as it currently does with /24 subnet limitation? I know that it does not limit the size directly, but slower you grow, smaller you can get medium-long term.
I mean, does storj want million of small operators (payout few dollars/month), or thousands of large, long time operators?
I want to have the opinion on the proposal, but I need to know, if I will have to stay smaller than 100 TB, or I can scale to multi PB at some time as storj scales.
I’m a bit sad to see the way some users are commenting this thread in an unpolite way. This is only a proposal at the current state and can be openly discussed - no reason to get rude here.
For my node the exact values are not that importan. A higher return is always better, that’s for sure. But my node runs on hardware, that is running anyways and the energy comes from my own roof. So my node would stay whatever will be the return in the future.
I do understand, though, that these numbers migth not be realistic for others. But then simply argument here for your position, say if this would still be an economic model for you to stay or not - but don’t accuse Storj for being “scam” and such.
For now the crypto market is bearish, but as soon as that situation changes. People will go to Chia (XCH). That’s for people who have medium to very large nodes.
Those who have some small node… as soon as they see that it does not even give them a sad tip… they will prefer to store movies.
Are you spending only 5% on SNO? I’m not wrong ?
And you decided to reduce them by about 4 times?
This is against the background of the fact that our expenses are increasing ALL.
You know, I started with the SIA project, and at some point it became more profitable to just turn off the equipment. Probably in STORJ this time has also come (
I would recommend you choose the upper limit of your proposed range, so $1 per TB storage, $5 per TB egress and $5 per TB repair.
At the moment I have 5 nodes, generating a revenue of approx. $50 per month. I observed that the egress per node declined over the past two years (although my total payout gradually increased). So if you cut the prices so heavily, it is a MUST that you generate a lot more customer traffic.
If you choose the lower bounds of your proposed range (and traffic remains that low as it currently is), I cannot guarantee that my nodes remain online. In Germany we have very high electricity costs.
For a lot of people, it will barely be paying for the electricity… not to mention tear and maintenance time… I don’t see if someone will even try to continue keeping up with the project if they will have to do it basically for free.
The best solution in my opinion is to reduce egress payout, for example to 5$ but SNOs should get 4x more egress traffic or something like that. In theory SNOs are getting paid less per TB, but overall they get the same amount. And that would be fine I think. So, in theory rates can be really bad, but as long as SNOs wil get nearly the same amount they used to, then it will be okay. Another thing is cold storage… $ per TB stored. I’m not sure how to solve this one.
Also, haven’t you thought about making some tiers for nodes? like flux nodes are? I’m not really expert here.
I feel like just reducing SNOs payouts to basically 20% is just bad experience for all people that invested their money and time. I thought you would maybe rearrange the rates a bit, maybe introduce some new way of getting extra buck to recompensate for the reducing payouts, but not to basically reduce payouts by ~75%.
And in my personal opinion something like this should be announced atleast year ago, not 2 months before
That’s very kind of you, but I have to say Storj is very open to community feedback. I recommend joining the twitter spaces and keep responding here to let your voice be heard. I definitely won’t stop advocating for SNO’s, but more voices sound louder and I don’t have the perspective of how these changes impact setups significantly different from mine.
In all fairness, Storj has been very clear that this was always their intention. The mantra to not buy anything for Storj is not new. They just had a payout model that made that clearly profitable, so many just ignored that advise. Including me. And as a result, there are parts of my setup that I would have to reconsider. Unfortunately for others with less efficient setups, it may be that their entire setup is no longer profitable, which really sucks.
Because these are the costs that will scale with network growth. If they can be profitable on unit economics, they can use that profit to outgrow all other costs that either don’t scale with growth or scale logarithmically with growth. They won’t instantly be profitable, but they open the door to profitability in the future, given enough growth.
This part sounds nice, but people don’t have endless storage space. The fact remains that when an HDD fills up, you can’t make as much of it as before. It is unlikely that egress will grow enough to even come close to compensate for that.
In a way, rising traffic will automatically happen when nodes leave the network. All that data will have to go somewhere and it’s going to the nodes that stay. That said, this won’t just increase egress per TB stored. So if your node is already full, there is no immediate benefit. Furthermore, even if they onboard new high egress use cases, full nodes will see little to no benefit from that as they are full of old data. And getting more high egress use cases is hard to begin with (and with the costly edge services, Storj may even currently have an incentive to keep egress low… which is really not where they should be). So I’m going to assume there won’t suddenly be a significant change in egress per TB stored.
For growth towards a sustainable situation, it’s better for them to let it happen naturally and listen to the feedback from both existing and new node operators. They can play with surge payouts to balance this transition as well as new node onboarding.
I think it is guaranteed. The new prices will simply make quite a few setups run at a loss. It may take a while for some node operators to notice this, but churn is kind of inevitable if setups start losing money. But I would agree that it’s really hard to determine the size of that churn. Just look at this topic though, there are already quite a few posts of people saying they will just leave.
I think there is a middle ground between not doing anything to limit the node population and blocking anyone from entering. Which is to limit the surge compensation to old existing nodes only. That way there is less incentive for new node operators to join and they could only join if their setup is profitable with the new payout structure. This would limit growth and also show Storj Labs whether they can still acquire nodes with the new payout structure, while existing nodes will be less impacted, at least at first.
This is the low range of estimates for internal power consumption. On average my hard drives use about 8.5W at the wall if you include power supply inefficiency.
Just wanted to add that these percentages match my calculations. And that lower range is mostly not survivable at my current energy costs if you have to pay for HDD energy costs (as in not used for other purposes).
Yeah, and from those 25% - 20% is energy cost
After you pay electricity, you’re left with 5% profit. Before you had 80% profit.
Which means the profit is going down by 1600%
That is not including taxes or other transfer fees
Bad pricing model: free tier without expiring which causes a load on the network without any profit.
Wrong target customers: put yourself in a competitive position with EC2, GCC, Azure… to get the big fish. Are you READY for this? Or you should get small business/individual professional?
No dev sponsors, not anything to get more developers on end-user products. Underestimate the normal customers.
Unnecessary low charge for bussiness customers to lower the barriers to entry: Why is $4/TB? Why? Your tech is not bad. Why charge so low? Your target customers are bussiness owners as you said. Why charge so low?
And now you want to pay less to SNOs, your providers, in order to cut the 5% spending on node operation. And a “proposal” cut of 75%? With that cut it’s basically zero profit or even a loss for most of SNOs.
This is the last choice and you guys want to do it at first.
Well, let’s include 20% tax crypto - so from 11$ you go to around 8$ then you pay 3$ expenses, so you are left with around 5$ per month. So you will get 60$ per year, assuming u’re using 8TB drive that cost ~150$, then ROI is 2.5 years, just for a drive.
Time + Wear is free I guess…
This is not possible, as those services requires to know encryption keys of the client in order to serve/accept unencrypted data. No customer keys can be handled by SNOs.
What can be done is to increase costs for customers using S3/linksharing services, as they incur in more bandwidth.
Aside from that, the lower values of new payments I think will put in risk the viability of the project in long term, with an icnrease of the costs of living (which includes the electricity) of ~5-10% this year (at least in Europe) and also lets remember that hard drives are mechanical machines that fails at some point.
Also, as @agapitox mentioned, lowering too much the price paid to SNOs will make other options more appealing for SNOs.
I get the point, it is not viable in the mid/long term to be spending 13$ of STORJ tokens for each TB of egress used (without taking into account edge services), so some adjustments has to be made.
I would prefer getting a higher amount per TB stored and less for egress, as egress doesn’t cost me anything directly and when the biggest part of the earnings comes from TB stored, the cashflow is predictable. Also for customers it might be more appealing to pay less for their traffic (costs are also better predictable). I like the idea to incentivize customers to use the uplink instead of the hosted gateway.
Just as a reference, when I calculate with the new upper and lower limit prices, my payouts would be 26% - 43% of what I am currently getting paid out. If I consider my current energy costs (which will stay the same or might even increase in near future), my profits will be only 14% - 33% of the current profit, which is not viable for me.
I just did math for my setup of nodes… (I calculated with the best 1:5:5 rates)
assuming my all nodes will be full, and fully vetted, I will earn about 72$ (72TB storage) + ~10% of storage as egress - 7.2TB(35$) = 107$, per month. Including tax and expenses into it, I will have around 25$ profit per month… which gives me a ROI of almost 8 years… just for the hardware… I don’t think I will live for 8 years to get that ROI back, so for me is a no go
I don’t think the majority of SNOs is on this forum. But it would be fair to let them have a chance to voice their opinion too. Otherwise you risk outrage and mass exodus when a new payout scheme goes into effect without notice for them.
Therefore I would suggest to inform the SNOs via node dashboard and email, maybe even Twitter about upcoming changes and this thread to take part in this discussion.
Let me take the chance to express my dissatisfaction with the multinode dashboard, which would be very helpful right now if it was working to check actual data numbers against your proposed payout. However it is broken:
It would be great if it could be fixed and the multinode app could become working and useful.
Maybe the node dashboard can have a section like the payment section, but calculated with the numbers of your proposal. So every SNO could easily see the effect of it.
