Are storj node operator rewars gona increase?

Node Operators
21

So, your node draws a lot for the storage it provides. Estimating from the income you probably offer around 12 TB of storage. 12 TB of storage is not supposed to draw 100 W. A single 20 TB drive draws around 10 W, plus some low-end computer that would charge another 10 W. No wonders you’re not efficient here.

Hosting a node is not just an exercise in providing storage, it’s an exercise in providing storage in an efficient way.

If you cannot justify running that PC for other purposes 24/7, then there’s no point in keeping it 24/7. On the other hand, if you have to run it 24/7 because of other things you run on it, then hosting a node there will at least recover the power costs.

Oh, indeed. I can imagine a SOC2 customer not trusting someone’s homelab :stuck_out_tongue:

6 Likes
22

“Hey IT manager! I saw a guy at Starbucks with a laptop and one of those external USB HDDs. Can I give him $5 to be an offsite backup for our 2023 financials? He seemed legit: it was a MacBook and everything.”

1 Like
23

So, your node draws a lot for the storage it provides. Estimating from the income you probably offer around 12 TB of storage. 12 TB of storage is not supposed to draw 100 W. A single 20 TB drive draws around 10 W, plus some low-end computer that would charge another 10 W. No wonders you’re not efficient here.

Yes, 12TB :slight_smile:
So imagine it’s RAID with 4 disks, and wattmeter plugged.
Sure, I could buy 16TB disk and some RPi on other arm board, and it would draw about 20W max.
But I won’t invest hoping ROI in few years.
But that isn’t point here. There are a lot of people who took the subject seriously and allocated a server with good hardware. Or even older PCs, that is not power efficient too.

Think about it, if you were to start a new node now, does that make sense? 2 years to fill your few TB of space? With hope that everything will be okay? IMO the good times are just about to end.

If you are in group that had some old RPi and bought big disk to backup their old films downloaded from torrents and discovered that they don’t need that disk, because they discovered that you can watch on streaming services - even better, you have some free money to buy good meal in restaurant once every month. Great for you.

How much does an employee cost an hour of your time?
How many hours did you spend with your node?
Is that time converted into money worth more than a year’s earnings from the node? Probably :slight_smile:

Putting it all together, if not today, soon being an SNO will make no sense at all.

Oh, indeed. I can imagine a SOC2 customer not trusting someone’s homelab :stuck_out_tongue:

Really? :slight_smile:
Do we all still remember that files on nodes are encrypted and have backup multiple times?
Do we all still remember that there is so much space left to fill on our nodes?
But no, let datacenters setup their commercial PB size nodes.

Anyway, it’s just my opinion, time will show…
Have nice evening :slight_smile:

24

The truth is, they shouldn’t be operating nodes then. Storj Inc. keeps reminding everyone not to invest in hardware just for Storj expecting profits. This ought to also include reviving old, inefficient hardware. If a prospective node operator with goal of profit does not perform basic economic calculus, then it’s their fault, not Storj’s.

Choosing to not operate a node altogether is also a valid move.

Too much, if you only consider profits. I’ve spent on my nodes maintenance probably some hundreds of hours over the more than 3 years, setting up hardware, preparing automation and working on occasional fixes. At my current contract rates, that’s already in tens of thousands of dollars worth of my time. I’ll probably never get it back from Storj Inc.

I’m operating a node for learning though, not for profits. From this perspective Storj has given me opportunity to learn tons of things, and the knowledge I got from it has already been quite useful for my work purposes. This might be a pretty unique position, but this is how it is for me. For others, who only want profit—follow Storj Inc. recommendations and if you do not already have hardware up and running with spare capacity, give up on operating a node. That’s also a valid option!

It still makes sense for many people to operate nodes for profit at current rates. My former workplace had a NAS with quite a lot of free disk space. I operated a node on that NAS while I was employed there, of course with permission of the business owner. It was pretty much maintenance-free. Storj Inc. currently aims the regular node operation program at those kinds of setups, and there’s enough of them to run the network at current growth… except for South America, apparently.

Regular old-school certifications do not make sense for many modern services. As long as they will be required for processing certain types of datasets, we can’t get rid of them though. It’s painful, sure, but if this effort gets us more customers that will also take advantage of the regular network, it’s good for us node operators.

At least we found that GDPR’s stance that good encryption is enough…

2 Likes
25

The truth is, they shouldn’t be operating nodes then. Storj Inc. keeps reminding everyone not to invest in hardware just for Storj expecting profits. This ought to also include reviving old, inefficient hardware. If a prospective node operator with goal of profit does not perform basic economic calculus, then it’s their fault, not Storj’s.

When you think about it, you can choose between RPi and old crappy disk on USB or keeping it at work, on your employer’s equipment with his knowledge or not. We excluded the rest before :slight_smile:

I’ve spent probably hundreds of hours too.
If it was only for profit, well… Like taking out a warm jacket at the beginning of next winter and finding money in your pocket :wink: A nice feeling.
But I learned a lot. I used rsync so many times moving node data, how to build all kinds of raid, arrays, lvm volumes and so on.

But when I read about big commercial nodes, only one thing can happen. Big guys will take all pieces of cake and SNO can only watch egress and repair data and lower and lower income, just to zero over time. Big guys with faster storage, lower latency.

Like you built the train tracks and when you want to get on the train you find you can’t afford the ticket. This project just leaves a bad taste.

I hope I’m wrong.

26

We had this discussion on the forum, I think, several times now. My belief is, when Storj becomes successful, big data centers will likely join indeed. But then there’s going to be enough traffic to also fill up the little guys as well, as with the /24 rule they’d still be offered a small slice of traffic.

I did hope that it will be possible to have a more professional setup for me in a nearby datacenter, but currently with the certification requirements, this is way out of my reach at this point.

However, the rules can and will change. The /24 rule is not set in stone, just as the requirements for a commercial node operator don’t have to always mean requiring certification. And so it all depends on Storj Inc. whether hobbyist operators will have a reason to operate a node.

27

To run Commercial nodes they must have SOC2 compliance, it’s needed for customers who requires SOC2 compliance. Very specific use cases. It’s not related to trust.
By the way the Storj network doesn’t requires trust at all.

28

My setup (10TB in total, ~$15/mo) is still profitable and pay my electricity and internet bills. This server will be online anyway, so I would still run nodes on it.

29

and do they still charge payments in tokens or dollars?

30

My server in parallel mines Monero to reduce power cost

1 Like