You can play with words all day long but in the end traffic costs money, api calls cost money, and storage cost money.
Providers that like to wow customers by pretending to offer one or more components for free just overcharge you for the remaining paid components. With R2 you just pay (significantly) more for storage.
I prefer to pay for what I use, not for what aggregate average of customers use + markup
On the other hand, “free” bandwidth, but more expensive storage is probably what I would choose for anything other than backups. The reason is that even thought the bills may be higher (not necessarily), they are more predictable. I would have better control on the amount of space used than on the bandwidth.
Just like renting a VDS - I pay a fixed amount for it and do not have to worry about using CPU time etc. Or my phone plan with minutes I usually do not use up.
You are correct, but that’s not necessarily a bad thing for consumers: egress is unpredictable, especially if you’re using Storj as an S3 alternative. With Cloudflare the storage is more expensive, the API calls are too but the egress is free, essentially meaning I’m just paying for what I store, and the amount of people that see it, this is significantly cheaper than Storj at scale with 100s of TB egress, it just doesn’t make any sense. The only benefit Storj brings to the table is decentralisation, as R2 also offers CDN and geographic replication. If Storj wants to remain competitive against R2, it would make sense to offer a similar product.
It may do that, but consider that the only difference between R2 and Storj currently is decentralisation. Which is great but in of itself it doesn’t make sense to pay in some cases x100 more per month. If you have a lot of egress (not necessarily applicable for everyone), which is usually the case for SaaS businesses. Outside of backups, Storj becomes unviable for any other use case.
Given all this, and the current economic climate - one can assume people would pick R2 over Storj, meaning Storj wouldn’t make any money anyway; unless they need decentralisation. The market Storj is in is very price sensitive as evident through out the post, so it’s a case of striking a balance with that to create a competitive product.
Even if what you say is true in no way does it make sense to increase the loss margin between income and payments. Storj make a loss right now and after they reduce payments to SNO’s there is going to be rather a large drop in income for those SNO’s - Dropping pricing further then means even more cuts for SNO’s which would pretty much almost free at that point or i again increasing the loss for Storj. Storj cannot exist if they make a loss in the long term - but neither can they expect SNO’s to be free. So, I do not see how further price cutting works in anyway.
Another 3.8M STORJ sent to Binance from STORJ MultiSig1 during last week. It is hard to discuss about SNO payments when you compare SNO payments (less than 0.5M STORJ in March) to these huge token movements.
So my employer has like many others moved their clients to cloud storage and for us the issue is NOT cheapest. It’s a bigger issue for us that Storj is a startup since many startups fail. We would rather pick an established business like Backblaze even if they are higher cost since we are less likely to have to move clients to something else in a mad rush if the business fails. There are no guarantees and no business is completely safe of course
You make great points, but perhaps it needs to be a combination of price cutting (but also increasing?) and an addition of new services:
Split Storj into multiple products: one for hot storage and the other for archive at different price points, thus allowing for it to be competitive for different use cases
Increase the price of storage to 0.005-0.015 per GB, this is a predictable income stream for Storj with minimal impact on consumers as it’s still very competitive
Offer “transfer” services, again similar to Backblaze to offer migration assistance, especially towards enterprises, now of course this is only scalable to a certain extent
Charge people for API access, like with Cloudflare while egress is free every time you load an object on the browser it counts against your API quota, this is still cheaper than egress but provides an additional income stream that is far more predictable to the end user and Storj, the pricing could be on par with Cloudflare potentially
The list could go on, but the idea is if Storj wants to compete in the market, it can still be a price leader just in a different way. This becomes ever so important as all providers start to offer the same value, and the only difference is pricing.
I do understand that SNOs need to be paid, and their importance in how Storj works, but I believe they can and should be subsidised by Storj expanding their product range and revised pricing, Im sure some middle ground could be achieved where all parties are satisfied.
This is a pattern across the industry, but that said the comments I’ve made are how things would play long into the future (3-5 years from now) where Storj isn’t a start up anymore. But for Storj to have a break through, especially in terms of profits, it needs to be competitive with its pricing, features and the total value provided to the end user.
If you visit Backblaze B2 landing page, you’ll see they immediately draw attention to their pricing, because that’s their go to, this is also similar to R2 where pricing is potentially the only deciding factor as most providers offer similar features anyway. The advantage that Storj has is decentralisation, so I don’t see why with a little infusion of new services and updated pricing Storj should be comfortably able to enjoy market dominance.
Maybe they are trying to move the price up, just to pay us less tokens? With these price manipulations everything it’s possible for the man with the biggest bag.
It’s not so decentralised any more with their gateway services.
The problem for them is that SNO costs have gone up with the rise in electricity prices which means the minimum node size that is now worthwhile has also increased. I used to build my nodes on 3TB drives. I would not do that any more. I think this does mean that the number of nodes in Europe are going to go down once pricing is adjusted - at the same time Storj doesn’t want to be dependent on places like Russia or Ukraine.
Australia is a good example. Very high electricity costs with another 30% expected this year.
Not a large number of nodes in the country. There is no way I could afford to run my current homelab there and it is one reason I left the country as I worked for a large energy consuming company at the time and saw the writing on the wall.
STORJ already offers transfer service and our solutions engineers personally assist important new customers in getting their data moved to our platform. Some of the other items you mention are being discussed. Stay tuned.
Right, as a consumer this is probably fine: you are essentially paying for the “peace of mind”. In reality there are just two possibilities – either you are overpaying, or underpaying (and are subsidized by other users who overpay), depending on your usage.
The former case is obviously undesirable. The latter – it’s a win if you are a consumer that uses more than average of all other consumers, but:
I would love a plan that would actually charge me for the CPU time I used, vs CPU time I could use. Because that would have been cheaper, you usually not running those CPUS above average utilization.
The phone plan is an interesting aside – I use 0 minutes per month, but about 120GB of traffic, from an unlimited plan. So yeah, I’m in the same boat, as a consumer. But I would not predicate my business on being able to get unlimited LTE traffic for a fixed price for long.
Why is unpredictable? It’s scaled compared to S3, but the scaling factor is more or less consistent.
I’m sure there is something in the terms of service that if your egress significantly exceeds stored amount you might end up needing to re-negotiate your contract with such providers. With Cloudflare it might be a bit more complicated, as being a CDN is their core business, but I would not expect to be able to pay $10 a month and egress petabytes of traffic
Even if I am overpaying in general, I am protected from surprise bills.
It may be cheaper by a little bit normally, but then something would lock up with 100% CPU usage and I would get an invoice for 20EUR instead of the normal, say, 5EUR. I’d rather pay 7EUR, but be certain that it is not going to be more than that.
It is unpredictable because it depends on the usage. Let’s say I run a video website (basically a clone of Youtube). Suddenly one video gets viral and that month egress is triple the average.
I’ve just asked on Cloudflare discord about this, heres a response from a community champion:
and here is their prompt on the landing page:
and I’m only highlighting R2 here, but you’d get the same result using B2 with cloudflare as the CDN, no egress. For context, the “Enterprise” plan is a bit expensive sure, which is between 1000 - 5000 USD per month, compared to paying the egress fees on Storj at 1 PB which is $7,000~ which becomes even more expensive when you consider the fact that Cloudflare provides a ton of other features in that enterprise plan outside just your storage requirements, that is if you needed it which is up for question.
To conclude with R2, you won’t have any bills related to egress, at most you’d have to upgrade to their enterprise plan which is fixed pricing. Your storage costs are higher but these are again fixed and predictable, you know exactly how much you’ll store, the only variable cost with R2 is how many visits your files get against their API calls, which is far easier to predict than with Storj, where you have to predict how many people will view your files x how big each file is. R2 just comes out cheaper currently. So outside of backups Storj falls short of anything that requires a ton of downloads: file sharing, image / audio / video hosting, even storing a camera feed that needs to be replayed on a daily basis will cost you dearly.
P.S. not affiliated with Cloudflare, even though it may look like i’m promoting them, I think R2 is a great product and something Storj can take inspiration from.
This thread reads like a flailing failing startup about to go under…
I hope not.
Also, I’ve seen quite a lot of ingress traffic on my nodes lately so it seems like there may have been quite a lot of new data so I’m hoping some big customers are coming on board.
Yeap, Amazon is closing their servers because the energy bill was huge last months, and they move all the data to Storj DCS.
