In the future, can we have an estimate of what % of the total traffic is due to testing ?
Like a short summary (with percentages if you can’t disclose actual numbers) at the end of each week/month.
There has been a lot of speculation on this matter and such information could really help clear things up.
As someone who stores actual live customer data on this network I think you should focus on performance rather than pricing. With thousands of nodes you can focus on IOPS performance and limits vs. throttling on Google Gsuite, Dropbox, Onedrive, and others. The sweet spot for customers with millions of small files is the current pricing but with unlimited and insanely high IOPS potential. The ability to transfer 1 million files worth 300GB in less than an hour is a valuable business proposition to most [insert favourite ecommerce platform here] owners.
As for the drop in traffic this is mainly because retrieval costs being too high puts STORJ as a backup target rather than a daily driver for your data wheelhouse.
You mean the past prices? What do you think about the new price scheme?
I was informed of the new prices ($7/TB) and for most small businesses this is an insanely good value. If I don’t have to wait 2-3 weeks for Google drive to download millions of files because it’s throttled this will allow me to launch more websites in a shorter amount of time and save on development staff time costs.
One of the use cases for storj is as an image file server for both backups and hosting. It can be part of a PIM/DAM solution. With millions of small files required to host an ecommerce storefront most of the offerings out there just can’t compete with Storj on IOPS. When you’re limiting factor in performance is your own local SSD’s IOPS limits you know you got a good DAM solution.
As long as Storj can deliver on both performance and reliability the pricing is fine as is.
Page loads mint new customers. The faster those images load the faster I make money. Google cares about page load times and will drop ranks on any store that can’t perform well on pagespeed scores.
7 Likes
@Alexey maybe this one belongs in announcements category?
was trying to find it… seems like an announcement… but i duno
This wasn’t an announcement but discussion of an announcement and how the new pricing would affect what storage node operators get paid.
1 Like
so maybe you can tell me what payout’s i should expect in just 6 months from now / 2022, because that seems kind of relevant pretty soon…
i don’t really feel or am aware of that that question was actually answered.
and this was the only post i could find on the forum about the new pricing.
but yeah i guess this isn’t actually an official announcement… just seems odd that one can’t find stuff like that… easily when it’s both recent and one of the biggest changes this year…
Yup, storj have not announced what if any proposed changes to payments might be made as yet apart from indicating they want to switch to a volume model. Apart from a belated concession they now realise zksync isn’t popular there hasn’t been much other news.
“Today we pay a premium for bandwidth to incentivize Storage Node participation in the network, but as we see use cases with higher bandwidth utilization, we may reduce the premium with the expectation that Storage Nodes will earn more based on volume. Our current pricing incentive is designed for that type of use case.”
From: Sharing Space for Fun and Profit—Part 2
If they do go that way - pity the smaller SNO’s…
yeah, well monday i’m going to ask for answers to that exact question.
we need to know what happens with our payout structure for 2022, to have an idea of what to expect…
i mean it doesn’t have to be super detailed… my setup has a lot of overhead, so it’s been challenging even to make it so it breaks even something which i’m still working towards, and if in 6 months the payments are slashed anything near what price reduction was, then i stand little chance of breaking even any time soon…
atleast ingress have been better, so there is that to be happy about, hopefully the worst is over.
but i think it should be a minimum of such things having a 6 month or such advanced warning so that we can have time to react and plan our upgrades or lack there of 
… maybe i should actually read that blog post…
had expected Storj to come out and announce something about what would happen with … lets call its SNO Payout structure, so that we have an idea of what to expect atleast in the near future, 6 months ahead doesn’t seem like to much to ask for…
1 Like
Storj seems to only take account of fees to get tokens into a wallet (be it L1 or L2) and then washes their hands of the situation but viability of an SNO is NET of all fees. Not just the ones storj decides to take account of and ignore all others.
I see two divisions happening in the project. Those who have been on board a long time and have been profitable (particularly due to the surge payments) and those who have joined in the last year. The second division I’d make are the smaller SNO’s with sub 3TB storage and those above. If you are sub 3TB AND joined in the last year - God help you! lol
well to be fair an avg of 3-4 $ pr TB which is about what we earn currently is actually pretty well paid… my main reason for not actually earning any money on running the node is that i run a big old dual socket HA storage server, which i sort of got for fun and to run storj on it… and i will keep it running most likely also if it doesn’t turn a profit…
ofc expecting a storagenode to make up for all that overhead, pay for internet power, hardware…
is a tall order, but doesn’t mean it isn’t possible… 
just need a lot of TB’s lol
and a fair payment for storjing them TB’s
i think what we is being paid currently is fairly good… 3-4$ is actually way over what i calculated for when i initially started of which i think was cold storage of 1½$ pr TB.
getting to the point where my setup is that efficient takes time, because it needs to store like a few hundred TB to get there.
I’m not disagreeing. I’m very, very close to seeing my first TB stored and before held back I am getting a shy over $3.00 per TB now. If you wanted to get tokens to fiat you’d still be waiting a while for it to be economically viable to do so at that payment level excluding other variables.
However, if the payment model changes - until we have details, who knows.
For one of my full nodes I see declining earnings and I am below $3 per TB for that one. The problem is that Saltlake is still the satellite that pays out the most, hence test data.
Actually you would need just a lot of egress. 1 TB stored that generates 20 GB egress per day would be so nice. Storj Labs really needs to find those high egress cases very soon.
2 Likes
well there are certain factors i think come into play…
-
the storagenode cannot be allowed to run out of space, because then the avg is disrupted, since all data will have a age range and not be an even distribution of data put into the network.
-
the node must work under near optimal conditions…
from what i can tell stuff like disk performance and network bandwidth is less of a factor than i would have expected, because i would sort of have expected the top tier customers to have different usage profiles than lower tier… so the last few % of the successrates might mean much… but doesn’t seem to maybe because we are mostly seeing test data or whatever i duno…
just know if i loose x amount of % successrates, then i losses x % of ingress.
i been mean to my node so it is also around the 3$ mark currently… but am trying to get it back to 4$… i think that the usual number is 4$ pr tb on avg…
but this will ofc change over time, and i guess the real data on it is kinda limit, there is a thread with node ages, tb and monthly earnings … can’t remember what it’s called… it also seems to correlate with the 4$ being the number for optimal conditions… ofc this will also be random… you might have better months and you may have worse months…
also i’m not sure how accurate the 4$ i forget how we arrived at it… so it might be 3.5 in reality and i just call it 4$ because thats what i believe…
i know i saw that for just about the first 6 months after my storagenode was 6 months old…
i suppose older data might also have a lower earnings pr TB due to the odds of it getting delete rather than downloaded will go up…
ofc it would also be a bit of a lottery because there might be large backups in that data which somebody needs.
personally my 16month 15.3TB node is still at about 3$ but has been declining overtime, but i’m pretty sure thats more my fault than the node… but maybe it didn’t affect it and i’m just seeing the same as you are while the node age…
its also why i try to fix it… to see if it is actually how it is or if it didn’t matter it shared an ip with other nodes for a long time
and still 3$ pr TBm is also quite good payment for storage… i mean price pr TB when you consider large scale deployment is pretty low…
I calculate the number in the earnings estimator. I just updated it yesterday based on recent behavior and it’s at about $3.20 right now. It has floated between $3 and $4 for a while now, but it does change based on egress and repair.
7 Likes
I’m new to Storj (yesterday) and just read this whole thread. Very interesting!
As author of HashBackup, initially released in 2009, I’m pretty familiar with the online storage landscape. As a relatively new entry, I believe Storj made a good move by undercutting Backblaze’s B2 pricing, because pricing is one of the first things a potential new customer will consider, and it has to be enticing for them to put in the effort to do further evaluation of the service.
I don’t know much about the Storj technology. It seems there are 3 differentiated use cases:
- backups: high ingress, high storage, lower IOPS (except rclone!), low egress
- image hosting: low ingress, medium storage, higher IOPS, high egress
- video streaming: low ingress, high storage, lower IOPS, high egress
Maybe Storj already does this, but it might be useful to be able to tag uploaded data to optimize storage node selection. For example, a node with a low outgoing bandwidth speed or limits and high storage capacity could easily host backups, but would not be very successful with images or videos.
2 Likes
I would recommend to make yourself familiar with the technology, and you would understand better, why Storj can do all of that right now without additional tagging.
https://storj.io/whitepaper
The only not implemented part is scale on demand (for popular files).
2 Likes
Is there an ETA for that feature?
1 Like