From the “Tardigrade is now Storj DCS” email:
We’re reducing storage fees from $10 per TB per month to $4.00, reducing our egress bandwidth fee from $45 per TB to $7.00, and getting rid of the per object fees.
How is this sustainable when you pay storage node operators more than you’re getting?
Good question. I expect SNOs will get an announcement soon on our pay rate.
While we will be experimenting with different pricing levels and models, we are committed to ensuring it remains economically viable to run a storage node.
While we’re modifying our pricing, we won’t be modifying how much we pay Node Operators. Node Operators can rest assured they’ll continue earning the same amount per TB of data stored and bandwidth utilized as they currently do through the rest of 2021.
I presume severely slashed pricing applies from 2022 on for SNOs. Don’t know if it will be worth it after that for us, but thanks for the explanation. Seems like there’s not a lot of adoption at current rate and Storj is trying something new. It is a correct step from Storj’s PoV, whether this suits us, SNOs, remains to be seen.
It depends on whether the traffic is going to increase to compensate for it.
If I get paid half the price for TB egress, but get three times as much egress, then it will be better or me.
If I get paid half the price for TB egress and still get exactly te same egress, then it won’t…
If you expect node operators to do more than just throw together any old hardware they have sitting around doing nothing you are going to have to be far clearer than ‘wait and see’. It is clear that you can not pay node operators $20 per TB for Egress traffic if you are charging the client $7 and even with the recent changes I can not see how $4 per TB stored can cover all the Storage, Repair and Audit payments going forwards.
Yes, but it has not been reduced by half, customer prices have been reduced by factor of 2.5 for storage and 6.5 for egress. There’s some minor differences but if you reduce the SNO payouts in the same manner, then it is 60 cents per TB stored and 3 bucks per TB egress fees. I expect the actual values to be slightly more optimistic, perhaps $5, if that, but it is not looking good. You will have to have a minimum of 5 times the traffic to earn the same amount, many SNOs don’t have the bandwidth for this. Personally, I have 10 Mbps upload, this means if I have the maximum utilization of 50%, average earnings over a year will not even pay for the drives’ power usage, let alone router and server power consumption.This essentially erases any and all profits.
dude i would bet there was some serious dough made from token price rise and storj can afford things or two now… lets say they dont have to earn on storage and bandwith in 1-2 years, and have money to pay SNO even without no revenue in 1-2 comming years how about that, can be possible, so they can put great offer and reach to the people
Sure, but as it is, there’s not much wiggle room with prices while remaining sustainable. They will have to reduce our prices, or increase theirs in the long run.
Decentralized storage is a great idea. Aggregating unused capacity from a distributed and decentralized network of storage nodes and making that unused capacity available to applications seems more efficient than building data centers, and it allows that aggregation of small providers to compete in a market dominated by hyperscalers.
It’s been a fun ride and a lot of hard work to get this far. We’re on the third generation of our network and we’ve delivered some really remarkable technology. With each iteration of the network, we strive to make a great product that delivers value to developers.
We’ve also iterated a number of times with the incentive model for storage node operators, from the per-node minimum in V2, to the standardized payouts, held amount, graceful exit and other incentives of V3. Here, too, with each iteration we’ve worked hard to make sure that the incentive structure is well balanced balance and ensures it’s economically viable and rewarding to be a storage node operator.
I know we’re not done iterating on the technology. There is too much great work left to do to make this a platform developers love. I also know we’re not done iterating on the incentive structure.
As we find better and better product market fit with the platform, it will become easier to align the roadmap and pricing with demand. As we continue to see growth, we’ll also continue to work on the incentive structure to ensure our node operators are also successful.
We have work planned on the product, on marketing and on the incentive structure. We get a lot of good ideas and feedback from the community on all of those things.
We’re also acutely aware that for us to be massively successful, it has to be economically rewarding to be a SNO.
So, yes, in the short term, we’ve made the conscious decision to ensure as we work through testing the market and driving adoption, we have a reliable supply of capacity by paying storage nodes disproportionately to the current pricing scheme.
Rather than cutting the payout, this gives us time to grow our customer base for Storj DCS and the time to continue iterating on the incentive model. There is a lot more to do, but we need our community of storage nodes to be successful if we are to achieve what we think is possible.
Ruskiem, unless they make statements there is no way to know. At the moment they could just cash in the tokens and pay out a dividend to their shareholders/investors, we have no ‘standing’/voice in their plans and so need them to be open about their plans.
If they are happy to subsidize us for the long term that would be great, but currently the only public statement covers just the next 8 months and 10 days. To put this in a simple context the renewal of my 350/30 cable service 2 days ago was for 18 months, as they gave me a very good deal I did not have to consider future income from Storj at any level, but other node operators may have to make harder calls.
At the moment the new pricing means I’ve now got possible projects that could use Storj as a storage platform, but I now also have doubts about the viability of the underlying storage solution as its future funding is not clear. This is a problem as any recommendation I make regarding using Storj as a storage platform will be a professional recommendation and so there can not be any doubt or clearly defined risk involved.
You are not being paid USD, thats how.
Hi storgeez.
jocelyn said: “we will be experimenting with different pricing levels and models”
I presume severely slashed pricing applies from 2022 on for SNOs
I doubt it - SNO’s are being paid from a large pool of tokens… Though the tokens arent’ mintable (I think they can be later on down the line). They are not monetary (e.g. not USD) either so it really doesnt matter how many tokens get paid out, there is no affordability really is there? It would be more costly to pay you in desert sand.
As long as too many tokens dont hit the markets, things should stay stable, right?
If many tokens go out and volumes on the markets increase, with lots of selling happening, then then price will come down, think OPEC economics, the countries regularly meet to discuss the amount of oil to be pumped to market. If lots of buying happens, the price will increase.
This is what I was getting at in my whitespace thread, which was just exploring the usage of unused resources to obtain more tokens, then I started to ask about the history of other models, it seems that something positive is on the horizon.
I previously fussed that it didn’t seem fair and things were under-utilised, it’s likely none of this action has come from my thread, it’s probably been in the maknig for a long time and im sure they will have more to come. (Unless they run out of tokens lol).
I need to fully explore this announcement anyway and even this thread.
It’s about time jocelyn, because when i was scoping this out wearing my (trigger warning) miners hat, I soon came to the conclusion that it currently makes no sense for me with my shiny new 10 TB external to get involved, and that being the case, it would be even worse to apply my infrastructure skills to get a cheap dedi and contribute to the project for a long-term investment.
They… do expect node operators to just throw together any old hardware they have sitting around doing nothing. Since inception this has been a project that uses capacity that’s paid for but otherwise idle, right?
With the new pricing for customers it may be profitable to run a storage node that stores its data on the Storj network :).
To be serious, I have enough unused resources that, to me, it would be OK if the payment per TB went down, but traffic went up more. Of course, if the traffic stayed the same (and payment went down), I would not be very happy.
I am being paid in a financial instrument that has a limited supply and a USD value.
The fact that Storj has a good few hundred million tokens available over the next few years and they have a lot of value due to them tracking the rise in Ethereum does don’t mean Storj has an unlimited supply or any intention to hand out more than it needs to. Until there is a published long term plan we have no idea what their views are. Also just because they have increased in value does not stop them from falling in value tomorrow, at which point Storj would be paying in USD as it would be purchasing tokens to pay node operators.
All we know for certain is that Storj has decided that it is time to undercut Backblaze in the s3 market place. Which I take as meaning that they have a lot of long term certainty regarding their customer-facing offering. We just need the same level of certainty regarding the structure of the backend.
This will also cause more data centralization, because it will be extremely unprofitable for small nodes to run themselves. They shouldn’t have gone under $5 for storage and $15 for egress as last resort. And I wouldn’t have been so aggressive with the first price cut. $7 storage and $25 downloads would’ve also been effective - less but still massively undercuts the big players. They can fix it later though by increasing the prices back up by some level after they gain adoptions, but it will be consdered asshole maneuvre by many and storage node operators with little bandwidth will be priced out of the market. Business is business though.
Centralization in your post would be more big snos, but the network doesn’t care about that. The data won’t get more centralized because of that.
As for small nodes not being profitable: this is just wrong. It may take you a while to see some profits but storj is designed to run on whatever you have running anyway. So if you run a Nas or a pi anyway, how can it not be profitable to run a node, however small? You might say it’s not worth your time and that is ok.
Your proposed price reduction would be rather irrelevant.
As long as the price is higher than backblaze, almost nobody would have switched. Now there is the option and many will (hopefully) make use of it.
IMO there has to be some amount of money earned that would make the time and effort worth it.
I mean, let’s say I could set up a node that would cost me nothing to run (free electricity and hardware). If the node made $1/month it would still be profit, but that $1 would not be worth it, especially considering transaction fees, exchange fees and such.
I think that quite a few people think like that, which means that if the earnings drop, nodes may start dropping off the network as well (probably not immediately, but people would get bored, forget to update or something).
Of course for operators who have more resources, the increase in traffic would hopefully make up for the lower per TB payments. So, it may start shifting toward fewer, but bigger nodes.