I want to see how it works on my RPI3+, but cannot revive it now… I’m far away from there and (wrongly) decided to keep OS on SD… So now it’s not boot. But since I do not want to return to there, I wouldn’t be able to test and see.
Could you please elaborate on the preferred location aspect? Do I understand correctly that nodes in Russia ultimately have less weight when receiving inbound traffic?
I do not think so. However, I see less downloads from my nodes (except repair), so this is an indication for me.
But as far as I know, we need more nodes in the South America region, see
Can you elaborate where your nodes are located? Whats your Energy-costs? Whats your Watt per TB? Whats your Running costs (€/TB)? 
Thx
Estonia (I like this country by the way, when I was there last time back in 2020).
I really wanted to live there, but… External circumstances voided my wishes…
@Bryanm I believe the real problem in your model is that you are comparing apples to oranges. Storj does not pay in dollars (fiat) anything to SNOs, you pay us in tokens. You collect revenue in dollars (fiat) which has to be enough to cover your costs of operations including developers, sales, hosting, overhead, etc. Every startup has a runway, you have one runway for cash and one for tokens. Devaluing SNOs makes no sense, we accept tokens because we have some faith in the product. Had you been paying us in cash, then there would be some sense in reducing cash payments, but instead as we accepted tokens, we have seen the those devalued over time.
I think it is not best topic to discuss that.
As the drive should be running anyway, some count the difference for idle and busy, wich is less.~half
Hello @Alexey 
I still don’t know how to quote in this forum system, so I’ll answer with topics, apologies in advance for that:
-
I have the stance that no income is pure profit. There is no such thing in this world as 100% profit. There are always things to consider; the PC, even a simple Pi, must run 24/7, the disk must run 24/7 or at least with some minimum uptime, a SATA is normally not that made for continuous operation like an enterprise SAS for example, the Internet line is paid (and used), and wear and tear are important if one is to for example get rid of the disk later on at secondary markets. I do agree though that investing purely for Storj only is not beneficial in ROI grounds; at 500GB per month, couldn’t be.
-
Thank you for the explanation on the long tail cancelation, I didn’t know that!
-
I would advise you to update your information regarding Sia hosting. Things have changed a great deal since the days of Siad…
-
Great point regarding some Storj strong selling aspects. Those are some of the reasons this project is essential.
-
Regarding the pricing per held TB, I agree. That’s why I’m proposing everything bundled together (changing pricing per held TB, egress/ingress, Reed-Solomon ratio and including the uptime and bandwidth criteria), in order for it to be sustainable. I left it more developed on the strike topic.
-
Disparities in countries do happen and always will happen. However, that is not the case in several countries: Portugal is just starting with 10 Gbps with Altice and Digi in the next months, but Spain, France, Belgium/Frankfurt and the UK already have some or most of it, as well as Romania. That’s where we’ll be in 12-24 months maximum, the uplinks already are giving 4-8 Gbps in Central Europe. In the US it’s pretty much a work in progress, with 10 Gbps guaranteed to some states and others severely lacking in that regard, but that kind of assymetry has always been a US brand throughout the last few years, so we’d just have to wait.
-
Thank you also for the explanation regarding satellites. But the expansion will be hard in South America at the proposed pricing, since the internal market suffers with extra import tariffs and substantially lower wages to account for using both currently-held HDDs, as well as current pricing for HDDs. That is also one of the reasons pricing should be adjusted, in my opinion. In Brazil, a change from 1.50$ to 2.25$, in a simple 4TB HDD, means that a node operator would recieve 29,85 BRL or 44.78 BRL for a disk to which they paid (included in a PC or bought as external for some backups of theirs, etc) at least 565 BRL. That is a significant difference in terms of value proposition. This is a best-case scenario and I can assure you that in most countries things will be far worse.
1 Like
Alexey!
Estonia doesn’t intrigue me personally, I’m wondering what’s wrong with the network, which according to you is very stable. What kind of corpses of nodes are these floating on the network?
3 Likes
More aggressive repair has been apparently kicked off anticipating future events. Nothing new around here, we’ve seen the same thing in the past.
1 Like
What events? I have never seen such an aggressive renovation before.
If this reaction from the storj to the planned action of node operators on November 11, then this does not look like a wise decision. It just might move the date to, say, November 1st.
But as I understood from the discussion, in principle, the storj said - do what you want, our network is stable and you will not affect anything on November 11th.
2 Likes
There was one at the beginning of the Ukraine war just to make sure the network would survive an entire country disconnecting from the global internet.
It is not. The repair work has been kicked off before the announcement was made.
Notwithstanding your other points, the UK absolutely does not have ubiquitous 10Gbps service.
In fact, for a large part of the country you’ll be lucky to get a FTTP service at all (although rollout is proceeding at a quick pace).
There are some limited 10Gbps services from smaller ISPs with a relatively small footprint but those are incredibly rare.
Can’t speak for the other countries, obviously.
1 Like
this is what most people cannot understand here. You have an online server, you already pay for it, include hardware and connectivity. But you have some free space on your disks. It’s not used, but drives are spinning to their death. Why do not share this unused space for some discount to your bills?
please update! I stopped to follow when I figured out, that if I lose my Metadata (SIA node), I’ll lose my data too, independently of the fact, that the data is “decentralized”. I do not care, if I cannot recover my files after my server is died. Really. Fully Decentralized? Yes! Can I recover my files if I still have my SIA wallet? - No! Why?!
this is interesting information, and worth to share with the team to consider, thanks!
meaning x2.75 for the customers. As I said earlier, this is not acceptable and makes a negative income. The best price would be $1, however, it’s not accepted by the Community (thanks for a feedback by the way!), so we have $1.5
I still do not see, how eliminating of egress payout to the Storage Node Operators would improve the reliability and the performance of the network. I still thinks that the payout for egress should remained.
2 Likes
Highlight text and click quote
2 Likes
You might want to poll SNOs here about your assumptions. I built a dedicated Storj node exclusively for Storj using spare parts. The parts are not ‘free’ (look up opportunity cost) and the power is not free. I do allow the node to use existing bandwidth. When I look at the thread on systems many ohers here built, often RPI’s those also appear to be exclusive to Storj.
2 Likes
We’re going around in circles.
Just because you chose to invest into buying the parts to spool a node doesn’t mean Storj should necessarily subsidise that, especially when they advise ad nauseam to only use equipment that you already have and would be running anyway.
7 Likes
I remember that for a long time the recommended way to run a node was to use a rapberry pi with a single hard drive for one node. I wonder how many people had raspberries with single drives attached to them used for other things…
I also remember debates about whether to use RAID or not. It does not matter then, because people would already be using or not using RAID when they set up their nodes and not redo their servers just to fit the needs of Storj, right?
Running the node adds constraints to how I can use my server. Uptime matters a lot more (Storj does not have such a thing as planned downtime), so does performance (when the filewalker runs, but in general as well, for example during a zfs scrub). Also, once the node has taken some space, it is difficult and/or slow to make it give some of it back. I cannot back up the node to try something risky on the server (I can back up everything else and restore if I have a problem)
Also, Right now, on my server, my node takes up about 20TB of space, or multiple hard drives (the drives in that server are 6x4TB and 6x6TB). While I use the array for other things, it could be made of fewer or smaller drives if the node was not there.
5 Likes