If you are not a hobbyist, you already have salaried IT staff. It costs nothing to plan for this “project”, and it does not take much to keep it running – storj keeps it running. Making sure container is alive does not add any workload to people who’s job responsibilities already include keeping an eye on other containers.
It it creates an extra non-negligible expense - obviously, don’t run the node. In the extreme, dont’ hire an IT consultant to help you setup and maintain the node. “Use what you have”.
I remember reading on this forum that storj targets to keep monthly payouts to node operators at about $130k. It’s not even a salary for ten entry level IT personnel. Of course paying someone to maintain nodes is not expected.
I do not see much of risk here, at least not from storagenode - it doesn’t execute what it has received and cannot do so, since it stores only one piece of the erasure stripes from a segment of the encrypted file. Basically it’s a digital noise, which is unable to harm anyhow.
But the misconfiguring a firewall (e.g. disabling it) may lead to a hack. This is a usual concern when you expose something to the internet and not related to Storj.
I would like to ask, did you consider to store your data on Storj instead of your own kind-of-DCs? We have a geofence feature, if that is a concern, so likely you can setup a bucket to do not allow the data to leave the specific geographic region (I understand, that even if it’s encrypted, there are could be some mandatory local regulations).
For the special software you can use cunoFS and that software even didn’t notice that is now use an encrypted distributed object storage.
There have been spikes every 12 hours (around 7am/7pm EST - with the pm spike consistently a bit larger) for weeks now. You can see it in Th3Van’s top two graphs (adjusted for his local timezone).
As I type this (around 6:30am EST) I’d expect to see the next one start in about half an hour. Maybe some sort of customer backup jobs (they’d have to be pretty big)?
We’ve verified the firewall configuration today through our CTO, and it has been confirmed to be secure without any issues. Additionally, our technical team also advised that there doesn’t seem to be any significant concerns, so we’ve decided to continue operating as we are.
As for using Storj as our data storage solution, we’re approaching it cautiously and observing the situation. While the cost of data storage with Storj is indeed much cheaper compared to other online services and has been on our radar since the beginning, we’re still hesitant, even with AWS.
The geofence feature is new to us, and it sounds promising—we’ll definitely discuss it further. We’ll also make sure to bring up cunoFS in tomorrow’s discussions. Thank you for the advice!
Ive been setting up a plan to build a NAS server that would heavily rely on decentralized storage like this and still possibly be profitable if youd like to look at it, ive been working on it for a few months now
I wonder what such a setup may look like with say a 8TB HDD? Maybe 2TB local storage and the other 6TB for a Storage-node/used-space (with that 3x ratio paying for 2TB Storj S3 to be the Cloud Backup for the local 2TB)?
I like the idea of offering enough local space to Storj… to pay for your own remote backups!
I’m setting up a NAS system for my personal use in home and wanted to dedicate some of its storage for Storj Node to hopefully cover the electricity cost of my NAS desktop PC. How viable is that? is it too optimistic for it to cover just the electricity cost of the NAS system? If not, how much storage I’d need in your opinion? I was thinking of 2TB but I guess it’s too low, right?
I have good internet connection 500mb upload and download and pay around €0.30/kWh of electricity (Germany)
If you know how much your NAS costs to run… it should be a simple to figure out. Storj pays ±$1.50/TB/month… so if your 2TB was filled you’d be sent around $3 in STORJ tokens every month.
It will take time to fill that 2TB, and time for you to receive full payouts from your new node: but it’s definately possible!
First of all; great idea. You’re helping provide a viable alternative to the large tech giants, you’re helping reduce the worlds carbon footprint, you’re probably going to learn a skill or two and you’re even going to get paid for it.
StorJ does not need thick provisioning, so I’d personally start with a single 1TB node, and scale it up as you go.
StorJ grows slowly but steadily - I see around 4TB per year for my sites. Will that be enough to cover electricity for you? I don’t know, but if the server is running any ways, than any non-zero payment will at least off-set the cost of electricity
Keep in mind, that the first year, you’re under the Held Back System, so during the first year you’ll see reduced revenue
[…]
Months 1-3: 75% of Storage Node revenue is withheld, 25% is paid to the Storage Node Operator
Months 4-6: 50% of Storage Node revenue is withheld, 50% is paid to the >Storage Node Operator
Months 7-9: 25% of Storage Node revenue is withheld, 75% is paid to the Storage Node Operator
Months 10-15: 100% of Storage Node revenue is paid to the storage node operator
Month 16: 50% of total withholdings are returned to Storage Node Operator, with the remaining 50% held until the node gracefully exits the network
The withholding model is designed to incentivize and reward both-long term reliable Storage Nodes as well as Nodes that, when they do choose to leave the network, exit in a way that is least damaging to the network.
Thanks @Ottetal for the reply.
indeed, any non-zero earnings is a bonus if the system is to operate anyway
good point to know about the Held Back System. I have a ton more to read apparently
No worries my friend. The forum is filled with a tonne of very, very well meaning people, who usually respond much faster and in much better depth than I.
Regarding the reading, then once again, the forum does offer a very deep iceberg of great swaths of knowledge, that can be worth reading. I’d imagine that advanced ZFS settings, filesystem optimizations and many of the other advanced topics are just not worth to dive into at this point.
The held back system comes as a surprise to some, but it’s pretty well put in the dashboard, once you get the node up and running.
Some simple break even math for you:
8.5 Watts (24/7 device use) x 24h x 365d / 12m / 1000(kwh) x .30EU->.35 USD = $2.17 USD / mo. cost
$2.17 / $1.49 (earnings per TB) = 1.46 TB
So, I’d guess after 10 months, you will probably break even for electrical cost; but not device utilization. By that time, it may be viable (considering trade/swap/transfer costs) to get STORJ crypto into fiat every 4-6 months or so, if you allocate more space (and STORJ uses it.)
Say after 18 months (25% device/TB depreciation) you have ~54 months of hdd use/life remaining. And may hold 6-7 TB / mo income, so ~$9.50/mo (<- likely optimistic)
All depends on your set-up quality and the customers’ utilization/location, the size of devices in your NAS you want to cover off (8.5 watts higher or lower per device), proliferation of Storj nodes, etc.
