Topic: Re: Fees for full nodes? - page 4. (Read 8464 times)

You can operate a full node* with pruning, what you're going on to describe there is UTXO commitments which are totally unneeded for pruning and are of unclear cost, unclear value, and unclear implications.


Unclear cost: rehashing the utxo set for every block, even if maintained in a high overhead efficient data structure is a multiplicative IO cost on running a full node.  Making the initial cost cheaper is essential, but at what runtime penalty is it acceptable?  is 20x acceptable?  If the utxo set is constantly changing, how do people synchronize it from you?  do you have to hold old versions? does a peer have to fetch it only from a single other peer?

Unclear value:  starting off a utxo commitment means you have only SPV security. If you were happy with SPV security; why not run SPV in the first place and dispense with the intermediate step. The answer is that it's only SPV of the history, but where is the dividing line? When someone today says how do you know that the system's creator didn't secretly mint himself a trillion coins-- the answer is because your own node verified it wasn't so.

Unclear implications: with no incentive to not run a committed bootstrapped node, is there any reason to believe the old history would even be _available_ to someone who wanted to do a full security initialization?

(*And in Git master Bitcoin Core relays blocks and transactions, and gives full wallet functionality, minus rescan when pruned.)

Currently, pruning is not the solution. You should be able to operate a full node with pruning. That means you need to be able to exchange UTXO information and an UTXO hash has to be stored in the block header.

I do think the discussion is interesting.  In the original description of Bitcoin, everyone who has a wallet is running a node.  Clearly this wasn't practical as the blockchain continues to scale and people wanted to run wallet software on their phones and cetera.  It would be cool if somehow the miners were tipping the nodes or something like that.  Perhaps this is akin to a waiter tipping out their busboy.  I dunno.  I guess I'm just here to follow along with the discussion for now.

I do believe in Bitcoin and want to see it flourish; I'm going to continue to run my full node despite no payment.  I was just hoping to get people to think about and discuss the topic.

I'm currently running a full node and it's taking approximately 1% CPU and approximately 80MB of RAM (obviously these numbers fluctuate).   My incentive for running it is that I want to strengthen the network.  This is more than just a "geeky good feeling".  If you're a part of a community, and you want that community to do well, and that community succeeds, then your efforts are rewarded.  I'm not saying that I don't want tips, I'm just saying that posting a topic to beg for a tip in return for keeping your node going is probably not going to result in a tip, at least I don't expect it to.  Good luck tho.

One is being developed. It may take a while to implement it, but every journey begins with a first step:

http://spreadcoin.info/news/proof-of-bitcoin-node-white-paper/

Then maybe in the future we would need something like bitnodes incetive program, but at a large scale.

A protocol level change to incentivize nodes would be very difficult, if even possible, for so many reasons.  Anything with incentive will be hit hard by scam attempts.  Also, consensus for changes becomes more difficult as bitcoin project grows.  You can be sure that any BIP to incentivize full nodes would be debated as much as changing block size limits has been.  I don't see it happening, personally.

FWIW, the bitnodes incentive program is now going to be continued past the original expiration date, thanks to 21.

Well, enable pruning and get rid of all the useless history!
(It's like cleaning your garage and throw away all the stuff you didn't use last 10 years)
Let someone else keep the full chain for future data archeology.

 
 
I was just about to make a topic about this when I saw this one. 
 
We absolutely need to compensate full nodes. To not do so is as foolish as to say "people will mine for the good of the network; no need to compensate miners".

I've been running a full node for quite a while now.  It runs me about $90/80 euros per year, it is usually in the top 10-15 on that leaderboard, although I am not sure if that is at all relevant or very useful. (4 Cores, 4GB RAM (5GB burst), 150GB RAID10, etc). Running a full node doesn't have to be expensive and it doesn't have to be difficult.  Just good to keep security updates applied regularly.

I don't believe that the fees would impact my decision whether to run it one way or another though.  Sure, if it was on my phone (give it 5 years and a phone might be able to handle it), but unless the fees are more significant, for me, they wouldn't play a role in me running or not running a node.  I'm sure not everyone would have the same opinion.

After 5 years, I may keep it going or I may not.  It isn't the cost, it is just a question of time to check it every so often and the lack of fees doesn't play a role.

(I doubt any fee for full nodes will happen any time soon from a consensus perspective anyway, though.)

 

my words^^

I can even provide IP publicly for everyone to check, that node is up, running and processing requests. point is, that nobody will give you anything, so amount of nodes will continue in decreasing trend + there is still aspect of increasing blockchain size.

I bet, that this will be sometimes subject of change, because serious financial network (tool) simply can't rely on geeks volunteering activities and good hearts of random internet strangers..

Transactions can contain any number of inputs and any numbers of outputs.  If the SPV node could request a bitcoin address from a full node then its transaction could include an output of a small amount to that address.  This would provide an incentive for this node to relay this transaction.  But between nodes transactions have to be relayed for free.  As it already is, nodes and clients must communicate whether they are full-nodes when they connect.  A client could lie though.

One solution is to run an Electrum server along with the bitcoin daemon and then leave a donation address in the server.  Electrum client users will see the address.  I know, you are begging here.  But at least there is an easy mechanism for people to support the node they will be sending their transaction to.

sdp

If a Bitcoin angel will give me 1 BTC now then I will run my full node as long as possible.

I already did it and recently switched to electrum, after running almost 2 years SSD based node with 100Mb/s symetric link. Simply, because there is no motivation to run it except some good geeky feeling, which is after years not enough.

the "Incentive Program" is just joke, which is based on luck and can't cover monthly expenses, even if you will win once every month...btw, I'm not talking about some massive fees, which can fund new ferrari, but just 5-10USD per month to cover basic costs..nothing else.

btw this is nothing new: http://www.coindesk.com/bitcoin-nodes-need/

I don't think the fee you receive will ever cover the full cost of running the full node. Running a node is "volunteer" work, you don't expect to be paid. I think Satoshi visioned most users in the future would use SPV wallets and only businesses or those who have financial interest in bitcoin to run full nodes.

I run a bitcoin node (ranked usually in the top 200 on the leaderboard: https://getaddr.bitnodes.io/nodes/leaderboard/ ).  But it is a bit of a beast and I often seen 400+ connections.  And the last non-stress test from coinwallet.eu when they released the free btc tripled the load on my server.

I also run a masternode on Dash (https://dashninja.pl/masternodes.html ), but that requires a 1K Dash investment for the masternode to have any remuneration.  I make about 2.5 Dash per 5 days.  That covers my cost for the Dash masternode and bitcoin node.  It is nice to get something back.

Read it before; read it again at your prompting.  I want to see Bitcoin succeed; would it kill us to pay full nodes a tiny fee?

I have pretty good internet speed; 90Mb/s down and 11Mb/s up.

I'm in bitnodes at 73.190.2.60 but haven't once received anything.

You are conflating issues. Requiring good bandwidth and large amounts of storage is a technical implementation issue. It is a barrier because it is not optimised rather than set in stone. Again, not financial. The current requirement for a full node is simply that they can verify transactions to provide consensus. They don't need miners' levels of processing power, electrical or hardware. The issue here is that the current implementation is onerous on bandwidth and storage for small devices. They are addressing the storage and one way to address the bandwidth is by spreading the load over more nodes. If every client is a node, each node needs less bandwidth. If you only have two nodes, then just those two are required to have sufficient bandwidth to do all transactions on the network within 10 minutes. The solution, again, is to make every client a node thus forcing all participants to support the network.