Q1) How fast do nodes receive and store data after being set up?
Q2) What is the expected profitability, considering factors like storage utilization, payout rates, and network demand?
From FAQ 
1 Like
This gets asked quite often: but if I had to give an answer to a friend, right now, Iād say:
- It would be reasonable to assume you could fill 3TB of space per year
- Given that youāll always be holding some trash, maybe estimate making $1.40usd/used-TB/month (after clearing the withholding period)
- Iād assume the first 1-2TB of used-space on each HDD simply goes to paying for itās monthly power+internet use. So if you assume on the high-end (first 2TB)ā¦ then a 4TB HDD could only ever make you 2 x $1.40 = $2.80usd/month profit. Soā¦ use the largest drives you can to avoid that overhead! (This fluctuates based on price-of-power and price-of-internet and efficiency-of-setupā¦)
But everyone here will have their own numbers. Good Luck!
2 Likes
ā¦ use the largest drive you have except probably not over 20TB or so because realistically a single node canāt fill up much more than that (max theoretical size based on bloom filters / garbage cleanup is 24TB i believe).
1 Like
Profitability was possible in the past when payment rates were much higher. A new node will likely never ROI.
3 Likes
- Depends on location. Around 3-5TB/year Iād say.
- Node payouts are 100% profit. The idea being is that they are supposed to be run to share existing underutilized, surplus capacity. An alternative to running a node is wasting that capacity and not being paid. Therefore whatever node pays is 100% profit ā as you donāt have any expenses, ROI is infinite.
If you plan to buy hardware and services just to run node ā donāt. If it was worth it ā storj would not need you. They could have done it themselves.
2 Likes
There is no maximum per se, but there is a equilibrium point between uploads and deletions, and itās dynamic.
For me, only systems with >7TB paid/month are making a profit, because the energy is so expensive. So at least 1 year to get there.
ROI? 
I donāt think about thatā¦ 
Iām going to suggest the opposite. I have 3-year-old nodes with less than 10TB and theyāre now stalled due to trash data deletion outpacing new data growth. Using HDDs larger than 10TB is a total waste in this scenario. You will never get ahead unless something change, you need client data to grow faster than the rate at which node operators are adding capacity.
5 Likes
This is on purpose ā itās a prolonged deletion of old free tier accounts. Perhaps storj decided to match it with growth to avoid payout drops to node operators. Itās a good thing. At some point all data from those orphaned accounts will get deleted and we will just get pure growth.
Ahead of what? You are still paid, no?
1 Like
Oh we are actually agreeing but you are making an even stronger version of the argument.
your (our) point: there is a maximum practical equilibrium per node. I currently have about 12 or 13TB data on one IP address at home (itās separate nodes and disks though, but one IP). So you can probably do bigger than 10TB, eventually, optimistically. It may depend on other vagaries like your physical location though at that point.
And then there is whatever maximum size is published on the storj system requirements. Thatās the max limit of the bloom filter, so if you ever did get more data than that one one node, it wouldnāt be able to effectively clean up trash and you would end up with excessive unpaid trash.
in hdd space soā¦ dont use 20tb hdd
I have currently 9 nodes behind the same IP with 35TB of combined storage capacity in central-east EU with residental internet access.
During the Great Test last summer, all of the nodes went full, but when the Great Garbage Collection ended, the combined used went down to 12TB.
Now the combined used capacity is at 17TB with around 900GB trash.
The max was almost 19TB around early January.
The nodesā ages are between 30 and 6 months and the whole thing gives me around 25-28USD monthly income. Electricity is free for me and I mostly use what I have, so I consider it pure profitā¦
I believe this 30ish USD is the max what I can get with a single IP.
I mean, if you have 20TB HDD-- use 20TB HDD. Why would you not ā free space is free space? Storj is unlikely to occupy all 20 ā but presumably you have your other files there, so you would limit storj allocated space anyway. Maybe I misunderstand what you mean. (i.e. if you mean ā donāt buy 20TB HDD just for storj ā then sure, but one should not buy disks for storj regardless of sizeā¦ so discussion is moot)
I know exactly what you mean and the message youāve been promoting for months. I agree with you about considering Storj as a side activity and using what already exists and is in operation to avoid having dedicated hardware (even if in the test season you would have been destroyed ).
In this specific case, Tarun used the word āprofitabilityā (the degree to which a business or activity yields profit or financial gain).
His request was much more specific and focused on business reasoning. He wasnāt asking, āI have a disk I use for other things, if I add Storj, can I earn some extra penny from it?ā. Soā¦ are we talking about profit?
Start with single node or multiple if you want to maximize ingress (If permitted by TOS), use an old, not SMR, 10tb hdd in the first 3 years. Minimize electricity usage. After that range of time you will understand if storj can fill up more space and will decide to upgrade or not.
Your node will grow gradually to the maximum of about 15usd/month at the end of third year.
4 Likes