Update Proposal for Storage Node Operators

Announcements
569

Well now that you mention it, As of before this proposal I didnt know about the edge services I thought gateway was just a direct way to upload and downloading and I still used my cpu to encrypt files.
But Now if the prices go up for this use I probably wouldnt use it and I wasted money on mountain Duck to be able to do this…But I wanted to support the devs who make our lives easier so I gave it a shot. If the prices for use of edge services go up there would be a reasonable price that id continue to use the software. I wish there were other ways to just mount S3 and rclone speeds are just horrible if not optimized theres no real way to really mount with the speeds that I would want.

Thats like me with buying the 4090 did I need it no did I want it yes so I got it.

570

Fair point. I just assumed that because it is an industry standard. I highly doubt that a developer would install uplink cli instead of using S3FS.
I don’t even know if “cloud sync task” from TrueNAS uses S3 or native. And I also don’t know where I would even have to go to tune that task.

I would argue most developers use S3FS for S3 and don’t host their own STORJ gateway. The fact that it is slower than a “consumer product” makes it even worse.
Which is fine, I no customer choses STORJ because max performance.

571

Or, at lest have a way to manually start some process to free up x GB of space.

How many people who buy these large drives plan on keeping them mostly empty for 3-5 years? Even if it was possible to shrink the node, you would need to have a lot of unused space for a long time (it takes a year for the held percentage to go down and it also takes a long time for the node to fill up).

Maybe this should change? It seems that developers are only interested in using the edge services, but a Storj-made backup program etc could be made to use the network properly.

572

Yes, I would be fine if they reduce my held amount for that. But having the option to shrink nodes would be really helpful in future.

1 Like
573

Hi, just a quick reply to your messages @deathlessdd, @Roberto, @Pentium100, @BrightSilence , a short one, because I am really on the move today.

I re-read the initial post of this thread. To my eye it looks like: a) there is a problem with a business model causing such drastic changes in storage node operators renumeration policy or b) to said it mildly somebody is trying to make jokes of storage node operators.

What I had on my mind when referring to the e-mail addresses and network growth including /24 rule is that I was proposing to close the network for new participants. This would result in a situation when one can create a new node only when participating earier and at least to some extent being an early contributor to the network creation.

In general, what Storj Inc. is proposing has significant repercussions for every storage node operator. With time there will simply be almost no incentive to run a node. There is no other option, this is how “a perfect competition model” and “commodity markers” behave.

Open market with a perfection competition model is extremely bad for storage node operators. For Storj Inc., well, I think it is not so good but I have to admit that to some extent it is discussable. Short term it is great because as I wrote earlier, the Company is in a business model of cutting a margin, so if a turnover grows, Storj’s Inc. revenue grows as well. However, again, please do think it over from your very own perspective.

To sum up, what is the future of storage node operators, what is the path forward? Are we going to upgrade our rigs to the level “proposed” by @Th3Van or maybe even to the level like this one or just keep it slowly running?

I am in a bit of a skeptical mood today but how you can’t be if you are reading about: i) results of free accounts promotion, ii) the product, including i.e. Python SDK, iii) /24 rule, including the ratio of nodes vs number of storage node operators and the whole Storj’s marketing story, iv) audited financial statement particularly the line titled “the other”.

I like computers and the community seems to be cool and vibrant so hope to stay around here at least for some time.

574

Hi John,

There could be many reasons that Storj needs to reduce this SNO payouts.
HOWEVER, we (the SNOs) are not really paid in Fiat currency. All we get is just storj tokens which has decreased in pricing significantly. Yet you guys are cutting the payouts means less and less storj tokens. That means we are doing it almost for free and paying for your company to have distributed storages.

If it is to discourage the sudden increase in SNO numbers…why don’t you stop giving authorization codes from that ‘Host a Node’ page. Then filter out bad nodes and shut them down.

3 Likes
575

I’ve just bought one. I planned to get one in half a year anyway for my lab, there was a promo yesterday, and if it still manages to earn some storj tokens before I’ll need to use it, all the better… though this is a case of an experienced SNO with an opportunity. Can’t imagine a random person joining fresh this way.

Anyway, when it will be installed, I’ll be at around 0.65 W/TB, going down from 0.8 W/TB. Yay!

S3FS is slow and its design as a FUSE file system does not allow it to avoid some problems. This
is visible even when connecting directly to AWS S3 from an AWS EC2 machine, i.e. the most favorable conditions. I’ve never seen it being used for anything but some lightweight dev work. Emulating a file system is hard work, if you can avoid it, you do so. S3 is usually integrated directly into devops or ETL tools, only then it’s solid. This is also I suggested Storj Inc. worked more on integrations. In my backyard that would be tools like Apache Camel, Apache Airflow, Django and so on, but there are tons of them.

The worrying thing though right now is that right now it’s not easy to write a compliant implementation without linking to golang and leveraging the existing libuplink code, and having to link to golang from non-golang environments is an obstacle that makes deployment troublesome.

576

Guys, here is the replay of todays meeting: https://twitter.com/i/spaces/1djGXlAqDOVGZ. This is potentially extremely cool Company, much better than Mr. Travis Kalanick early startup. The management and technical skills are absolutely there. The automation of selling process should be implemented, in particular the product experience should be improved thanks to a focus on end user and developer experience (please see tailscale.com which has a much more difficult product that is extremely easy to use), individual sales airplane experience only in cases when securing a long-term clients with a minimum of at least 4 full 42u racks achievable. Units’ economy of individual storage node operators should be increased and /24 rule gradually removed (its only on the paper currently and everything works great), particular focus on improving the skills and machines capacity to be implemented (please see the programs and tactics of uber.com). Object storage market disruption strategy should be redesigned and two to three, the most promising Web 2.0 / Web 3.0 market segments studied, with product and communication strategy implemented.

1 Like
577

Nodes are paid in Storj token that is fixed to USD pricing at payout time. Flux in the market is common, nobody complains when the token prices rise, just when they go down. A lot of token value is derived from where the entire crypto market is, and that is largely driven by speculation and other markets.

3 Likes
578

Are you sending any info when the payment is made? I have never got any e-mail on this topic from you. It’s a bit difficult to implement any strategy with regard to token volatility.

1 Like
579

That’s far off topic. If you want answers to questions that are off topic, I’d first do a search because this question and how those calculations are done at payout time have been answered numerous times. If you still have questions, please create a new topic.

581

So I feel much better after the Q&A. Seems like Storj really listened to the community here and are taking many of the suggestions into consideration. I’m hopeful that a good balance can be found between cutting SNO costs as well as finding ways to increase costs to customers. I really think something like a tiered structure with higher cost per TB but low bandwidth cost would be very beneficial for targeting content distribution type use cases and hope to see this sort of thing being discussed further.

I’m especially excited at the prospect of using collateral in order to remove the subnet limitations in order to host larger nodes. I believe this would help nodes to scale in a way where the reduced pricing can be offset by higher efficiency and the capacity for higher egress bandwidth utilization, as well as eliminating the costs to large SNOs currently associated with circumventing these limitations. There’s no sense paying more for the same (if not slightly worse) end result. On the other side of this, I think it’s also important to keep the current model of holding funds as this makes for a nice and easy no risk barrier to entry for new SNOs just getting started.

4 Likes
583

Let’s not act unprofessional and make troll posts like this. I want to allow everyone to say what they feel about the payout proposal, but stay on topic and don’t make bashing comments unless they are constructive. I would prefer not to delete responses because we want everyone’s input. But we will if it veers too far into being off topic or bashing.

4 Likes
584

Nodes are paid in Storj token that is fixed to USD pricing at payout time.

That I know.

Flux in the market is common, nobody complains when the token prices rise, just when they go down. A lot of token value is derived from where the entire crypto market is, and that is largely driven by speculation and other markets.

I am not complaining about the crypto price drop. Where are you going with this?

My comment was for the proposal of reducing payouts for the SNOs. It’s not just reduction, basically they are dropping it to nothing. The SNOs are paid in storj token. how much? that is calculated based on Fiat currency.

However, who generates storj token? They are not buying these storj tokens using their fiat currency, isn’t it?

Crypto tokens can be and are being generated based on transactions in the network. So, technically the SNOs are vital part of this network and they are proposing to reduce the pricing for egress from $20/TB to ($1.5-$5). This is insane.

If you simply reduce the synthetic data from the network it will automatically reduce the payouts. Thus it will discourage the recent increase in SNO numbers.
That’s all.

585

All the tokens were produced years ago. Storj does not create new tokens.

I think you are under the impression that SNO’s somehow mine tokens. They do not. Storj has tokens in reserve and pays SNO’s with those tokens. When they run out of tokens, it is assumed they will purchase tokens from the market for distribution to SNO’s.

Anything further on this is off topic. Please create a new post if you have questions on the tokens.

2 Likes
586

Thank you so much for enlightening me on this. However, it is still true that,

the SNOs are vital part of this network and they (the management) are proposing to reduce the pricing for egress from $20/TB to ($1.5-$5). This is insane.

If you simply reduce the synthetic data from the network it will automatically reduce the payouts. Thus it will discourage the recent increase in SNO numbers.

This is the topic behind my replies.

587

I think we need to reach an agreement on lowering the amounts paid to SNOs, for the good of Storj and the SNOs.
I suggest that Storj strike a balance between what it charges its customers and what it pays to SNOs, at least for now.
A three-step plan, each three months apart, could be possible.

  1. Storage (per TB Mo) $1.50
    Egress (per TB) $15
    Audit / Repair traffic (per TB) $8

  2. Storage (per TB Mo) $1.50
    Egress (per TB) $10
    Audit / Repair traffic (per TB) $6

  3. Storage (per TB Mo) $1.50
    Egress (per TB) $7
    Audit / Repair traffic (per TB) $5

In return, Storj must do everything in its power to reduce its other costs. But we have to keep in mind that there will certainly be further revenue cuts for SNOs.

588

Abstracting from all the business topics. Thats how you presented the proposal so you should not be surprised by the reaction of the readers. Apparently all this, especially your behavior, does not look like it is about friendship. You deleted my previous post thus with this one I am copying your supervisor so he will be aware. [@john]

Very low of you guys.

Edit after the post hidden by the “Community Moderators”: Nothing to be changed here. Please see: Update Proposal for Storage Node Operators - #599 by s-t-o-r-j-user particularly the second paragraph. Cheers.

3 Likes
589

The one where all you wrote was, “Very far, are we still friends?”

This thread is for discussing the payout proposal. If you’d like to start a new thread with your question, please do.

590

I have mixed feelings after the twitter spaces. Great to see that S3 is only a fraction of the costs and I feel you guys have not given up. Focusing the marketing on web3 is risky in my opinion and the spending of “others” was left unanswered.

For me it wasn’t convincing enough to spin up a node. I will reevaluate the situation in 6 months.
See you guys! Have fun and remember to only gamble what you can afford to lose.

2 Likes