No. In case of Graceful Exit your node will transfer its pieces to other nodes, so repair is not required, it will just move a piece and update the pointer in the metadata.
3 Likes
No. If you do a graceful exit, then no repair is required, this is why the held back amount is returned in this case.
Edit: @Alexey beat me.
This is why incentive is proposed for nodes staying long time on the network and nodes doing graceful exit.
2 Likes
You need to take into account, that a segment not necessarily needs to be repaired.
Repair is only required when the number of pieces falls below the threshold.
1 Like
Well that kind of cancels out. The lowest availability is currently 52, so 28 pieces need to be recreated. Your 1 lost node is responsible for 1/28th of the cause that the segment needs repair and in order to repair it, the repair worker needs to download 29 pieces. So itās pretty much 1 to 1.
1 Like
I believe these days ingress/node is more than what you calculate with. (Yes, I know, long term statsā¦ but maybe it would worth to tweak the numbers again)
I had a 2TB HDD laying around and I started a new node with it by the end of August.
This node started to fill up quite quickly, so by the end of October, I started an other oneā¦
On the first mentioned node I have around 400GB free space left as of today, so so after 3 months it stores 1,3TB of data.
On the second node, in 3 weeks, I have 175GB data stored.
I think that the 2TB drive fill very quickly. Most probably the held % will be still at 50% when it will be fullā¦ This means to me that it does not worth to start a node with less than 4TBā¦ which lead to the point that average users/node operators can hardly start a node on leftover hardware (just to have some connection to the topic as well)
Iām seeing about 800GB ingress + 100GB repair ingress in the last 2 months. But I think the difference you see compared to the earnings estimator is more because the loss to vetting is actually less than calculated. I adjusted both numbers a little based on some new nodes I started myself. It now seems to match what you saw.
(and yeah, I adjusted it a little faster than I normally do, because I prefer waiting for at least 3 months to see a trend reliably change, but I think the past month and a half combined with Storjās statements around growth probably present a reliable enough basis to update.)
Let me know if it looks better to you now.
Source: Realistic earnings estimator
1 Like
Yes, the vetting process is much faster than before. My new nodes became vetted for EU1 and US1 in about 2 weeks.
I checked the ingress across all my nodes for this month and it is already over 747 GB even though we are only halfwayā¦ they are all behind the same IP.
What is your opinion about the node size?
Maybe it would worth a calculation with 2-4-6TB node size and the revenue expectation considering the decreased egress rate for full nodes.
In october, my 3,9TB full node had a bit more than 9 USD revenue. This node is full since June:
Also in october, my other 3,7TB, also full node had 14 USD revenue. This node became full around the end of Sept:
4ā¬/month?
What drive is that?
How much do you pay for kWh?
I suspected that is how it should work. The problem is it also means the graceful exit makes no sense. Let me give you an example with real data. I have 6.6TB on my oldest node (more than 15 months) with 6.46$ as held amount. If I would exit gracefully, it would take 1 month at 25Mb/s to upload my node to other nodes, and I would get 6.46$ for it. Now, if I can wait 1 month before switching off my node, then I would rather run it as usual for that month and ungracefully switch it off by the end of one month. By doing that I would get 20 plus $'s, while not having my network running 25 Mb/s upload for a whole month.
2 Likes
I agree, at the moment held amount is basically broken. Itās been suggested several times to just keep building held amount until it amounts to $10 per TB. In your case that would now be $66 and probably be well worth it to do a graceful exit. But as you mentioned, currently itās cheaper to just run the node until you really need the space for something else and then just kill it.
3 Likes
Itās a 3.5" external drive which uses around 10w. I pay around 0.5ā¬ per kWh.
I seeā¦ well, the 10W figure is normal, but 0.5ā¬/kWh is overkillā¦
Maybe a solar panel? You get more than 10 hours of effective daylightā¦
Just for the disk would not be worth it. But consider everything you have running 24/7 (computers, routers, switches, some ill thought infrared receiversā¦).
~350W panel is ~250ā¬
microinverter is ~140ā¬
differential circuit breaker is ~30ā¬
instalation materials varie a lotā¦
Even if the sun is not shining it still should output ~50-100W, as long as there is daylight.
Depending on the country you live in, you might get away with just connecting the microinverter to your power network and forget about it.
Depending on the country you live in, you might have the right to install it on the roof even if you live in an apartment.
This will cut your 24/7 low power appliances by ~50%.
yep, 66$ would be worth a graceful exitā¦
But would it be worth for a starting node operator to invest his time and, in some cases, money, to wait for so long (1 year?) until he sees some tokens?
Also, like the current held amount system, that would be a tax on serious node operators that have no intention of leaving the network, just a bigger taxā¦
I can not agree with a system that effectively cuts the earnings it promises to the node operators, whether 6.46$ or 66$. It kind of looks like a similar ongoing business that is based on the model of ānot paying the node operatorsāā¦
That said, since this thread (which I knew was coming, but hoped it would come laterā¦) will be used by the storj staff to feel out what they can count on from their āworkersā (SNOās), I will leave my 2 cents on the subject:
I obviously want storj to succeed and prosper, but itās clear the current model (prices/payments) is not sustainable. I think itās clever to subsidise the network under the prospect of substantial growth (itās just tokens anyway, though the circulating supply should not grow faster than the network itself). Itās also obvious the āworkers for freeā model is not sustainable either. It doesnāt matter if only 10% of us say they will leave or if 50% say they will leave in case of severe cuts. In the medium/long term we will all do something else with our disk space if it is not profitable, then the project will just die.
Most of the numbers discussed so far for SNO payments canāt really cut it (one SNO mentioned paying 4$/month on electricity to run 1 external disk, meaning, if it is a 2TB old disk he had lying around waiting to be discarded, heās basically being paid to connect the disk to the power grid). That is not sustainable!!!
The way I see it, storj can not compete on lower prices alone. It has to bring added value to the storage market. It has to offer something different and find its niche (CDN is a great argument, although first byte speed is not so good). Also, as someone (sorry, canāt remember the name) has pointed out, it doesnāt matter how technically good you are, without good marketing youāre dead.
Storj should grow until it finds its size, even at the cost of subsidies. It should aim to become a player in the storage market. It has a great defence against being crushed by other big players, which is the use of tokens. The consequent release of frozen tokens is bad only when it stops growing. Meanwhile it should technically focus on what it can bring different and/or better to the market. When it reaches its niche and stops growing, then we should address this big heavy elephant. By then, hopefully, the market will be different and weāll have a better perspective.
On a more personal basis I would say this: I will not subsidise the network running unprofitable nodes (duh!). Since Iām not running anything specifically for storj, I will not switch off my nodes just because itās no longer profitable. I will switch them off when I need the space for something else (duh again!).
While building to this amount only 50% should be held back. So the operator still gets paid, just less.
Well there is no sign from Store Labs of any specific changes coming. I may have jumped the gun a little, but I thought it would be good to get the conversation started.
Energy prices in Europe are crazy right now. It wonāt stay that way forever though. But this is probably not the right time to lower payouts. I live in the Netherlands and unfortunately prices are similar for me. But starting next year their will be a cap to how much energy suppliers can charge. And weāre already getting compensation for the last months of the year. Weāre in an energy transition that is forced to speed up significantly because of geopolitical strains. Weāll get out the other end of that some day.
Absolutely, but they need to have more market recognition first. CDN cases work great for larger files already, where the time to first byte is less critical. Cases like software distribution or video or large data sets.
I think you hit the right points. The timing is bad. But keep in mind that it was me who started this discussion, not Storj Labs. And theyāve made it clear they have no intention to āgrind node operators to a pulpā. Gotta love that quote, haha.
Sure. But the point still remains. If you have made a few thousand already, the 66$ held (forever!!!) by storj doesnāt make much of a difference. But try to explain a new SNO that the promised 1.5$/20$ will only be effective in 1 or 2 yearsā¦
And you did good. Iām not criticizing the initiative. It was coming and you just had more guts to start it. I do have a few problems on discussing actual numbers (for paying SNOās) at this stage. Nobody is getting rich, not even Vadim (whose setup I donāt understand, BTW!). But throwing in the air numbers like 0.73$/5$ is laughable. Maybe some older big SNOās can cope with those numbers. No small or new SNO (math conscient) will run a node.
Iām not sure we will ever have low cost energy again. We will need the high cost to make the transition (forget about Russia invading Ukraine). And that may not be a bad thingā¦ in the great scheme of thingsā¦
Yes, I think that is THE point. That is why Iām so bullish on growing, even if subsidized growing. Storj is in a great position to grow because it doesnāt need investors to grow. Growing does not bring costs (they donāt have to buy more facilities, more disks or pay for more energy or infrastructure), and even if it did, they have already printed their own money, which also makes them more resilient when facing ācompetitor attacksā. Not to aim at growing under these circumstances means either you are short-sighted or the world just doesnāt need more storage.
First, letās buy a sit at the table, then weāll play our cards.
I was never worried for a single second they would grind me to a pulp by cutting my ālamb buyingā profits. They need SNOās as much as they need clients. Exactly as muchā¦
You know what would be stupid? To let the project die while sitting on ~200M tokensā¦
The āgrinding node operators to a pulpā was a joke @john made in response to me addressing the discussion in this topic during the last Twitter space. Itās worth listening to his response, because I think he echoes exactly the sentiment youāre expressing. My question starts at around the 20 minute mark.
https://twitter.com/storj/status/1588607064955588608
Itās about a 10 minute response that while not containing specifics, does very clearly state Storjās attitude towards the current situation and goals.
Tomorrow there is another Twitter spaces planned if you want to ask your own followup questions. Unfortunately I canāt make it this time, but it would be awesome if other node operators could chime in.
Iāll be sure to listen back afterwards.
That was the whole point in mentioning numbers. To make that response very clear. And I think in balance this topic shows that $1/$5 is not acceptable for too many nodes.
I can make some calculations to show what it would actually look like. But you would only have to build up to that $66 while your node grows. Most months you wouldnāt have the full 50% held back because your node doesnāt yet have to build more collateral. Let me get back to that in a bit. But I see your point. I would like to say that some other networks have you stake tokens upfront, which is much more of a hurdle. I like that storj doesnāt require that to begin with and just uses held back payouts. It would also be possible to lower this requirement for long term reliable nodes. Say pay back half after a year of running your node and only require $5 per TB after that. These knobs can be tweaked so it makes sense for everyone.
It would look something like this for a node that keeps growing.
Yeah, I think some fine tuning is needed. But this gives some idea of what I was thinking of.
For a 5TB node it would look like this.
Would a couple month āgrace periodā delay to the held amount be helpful to new nodes? Having a larger cut of the pie taken when your earnings are so low seems like a bigger hit than waiting several months for the earnings to grow a bit before taking any for collateral. I realize this could increase the amount of new nodes leaving due to lack of punishment, but could it boost the perceived payment for new nodes, incentivizing new SNOs to stick around more? It just seems like a dial to move to find the best option.
P.S. I guess the basis of my thought is, where does storj want to prevent churn? On new nodes, or older nodes? So if churn at 1-3 months doesnt matter, does āincreasingā payout at that time increase the number of nodes that actually do stick around, where you can take more collateral, while its less to the SNO.
Thatās far too high for a 3.5" HDD. From what I read 6W would be more realistic.
My server draws ~125W on average and has 9 HDDās and 1 SSD. If the 10W would be true then motherboard, CPU, RAM and all the fans would use next to nothing.
It ranges from 5-9W, but an external USB disk adds additional overhead due to the power adapter loss of efficiency. 10W sounds about right to me.