If Running a Bitcoin Node Became Too Expensive, Would Decentralization Suffer

[Gost ms]

Yes, it will definitely be, and it will probably never happen. Because if that amount of electricity is the price, then at that time about 50% of the people will not have electricity in their homes. So the possibility of such a situation is much less. Whenever there is a shortage of electricity, there are many means through which electricity can be generated and the shortage of electricity can be dealt with.

And if such a situation ever arises, then the price of Bitcoin will definitely increase a lot at that time. When the price of Bitcoin increases a lot, a person will get a lot of money by verifying transactions, with which he will be able to pay the electricity bill. If the number of nodes running is less, then Bitcoin will not be centralized, but rather the possibility of a 51% attack may increase.

[Leon Zimmer]

shows since 2023 many blocks almost fully reach block size weight limit. So from that angle, it won't become worse.

Fair point. If blocks already sit near the cap, the limit is what bounds node costs, whatever the content. So this supports defending the weight limit rather than being a separate risk.

Do tell me how a node can prevent a miner from censoring and invalidating transactions!

It cant, and I overstated that earlier. Nodes enforce validity, not inclusion. What they stop is miners rewriting the rules (inflating supply, spending without a signature, etc.). Censorship of transactions is a separate problem that nodes dont solve.

So, can you please elaborate next time the "costs" of running a node if we move to let's say 10vMB blocks?

10 MB*144 blocks*365 days is about 525 GB per year (assuming blocks are always full). That would cost around $50–80 per year using SSD storage at today’s prices.

Bandwidth: about 1.4 GB per day to download, and around 350 GB a month to upload (assuming 8 peers on average). For an unmetered connection that’s a few beers a month. On capped or metered connection it can eat up the whole data allowance. Thats the group I meant, not average German.


I think we have covered the main points. Nodes enforce validity, not inclusion, and cost of verifying is still real tradeoff if blocks get bigger. I will leave it there.

[Nathrixxx]

If only wealthy individuals, large companies could afford to run nodes, would Bitcoin still be decentralized as it is today?

What's determines for Bitcoin decentralization is not in the ability to afford to running a node, the only requirements to this is when you can afford the device that could help you download the full blockchain, and the benefits is for privacy purpose.

When we talk about the centralization, you don't have to mix it up with privacy because the blockchain technology operates as an independent network and everything is programmed under the protocol that only permit for full compliance, anything not in consensus will not be accepted, there are other ways you could also maintain decentralization and privacy like using a decentralized exchange, using bitcoincore you are so on.

[stompix]

So, can you please elaborate next time the "costs" of running a node if we move to let's say 10vMB blocks?

10 MB*144 blocks*365 days is about 525 GB per year (assuming blocks are always full). That would cost around $50–80 per year using SSD storage at today’s prices.

Bandwidth: about 1.4 GB per day to download, and around 350 GB a month to upload (assuming 8 peers on average). For an unmetered connection that’s a few beers a month. On capped or metered connection it can eat up the whole data allowance. Thats the group I meant, not average German.

Good, so finally we have some numbers!
I'll ignore the metered connection because it's 2026...like f shit I have unlimited mobile data for 15 euros, common!

So straight to the thing...
It will cost users $80 a year to keep a 10vMB block running.
Making a transaction at 1 sat/vb versus 2 sat/vb saves you 10 cents per tx, if you make a transaction each day this will be $36! Oh wait, there is change also, so you will have to consolidate once your change and for ~300 that is 20vkb which at 1 sat /vb will also save you another $16.
Now that we have gone over the fact that an actual user would basically get his money back just because he spends 1sat/vb less on a tx let's see the second part.

$80 a year would be 4 ?? Maß 
Sorry, but if we can't find 50k -100k people who are willing to spend the price of 4 beers at Oktoberfest or the price of GTA 6 to challenge the old system, then the new system has failed!

Oh, did I mention that if we were to migrate to LN it will take 30 years for eveyone in the world to open a channel?

[Leon Zimmer]

Now that we have gone over the fact that an actual user would basically get his money back just because he spends 1sat/vb less on a tx

Putting aside the metered connections problem, fees are not really an issue. They are already as low as they can be (1-3 sat/vb) as it is and bigger blocks won't really help bring them down much further. If you actually do save money, its only when there's congestion and only benefits those who do a lot of transactions.

if we can't find 50k -100k people who are willing to spend the price of 4 beers at Oktoberfest [...] then the new system has failed!

Validation isnt free, and I dont think $80 is outrageous. What I m saying is that there are trade-offs to validation that we should consider.
$80 is only the cost of storing the chain at 10 MB. There is also the cost of increased sync times and validation load that scale with chain size. And once we start moving the 10 MB number, it will probably never stay static.

[KiaKia]

Relevant costs are the ones which the Bitcoin protocol can control: bandwidth, disk space and CPU. This is why its important to defend the block weight limit, because if it was increased a lot then its likely that only datacenters could run validating nodes, as you describe.

Oh god, just put a fork in it!

Seriously, sometimes I can't understand the way some bitcoiners think.
One moment is bragging how one Bitcoin would allow you to rule from your castle the fiat plebs, then it's screaming about how they can't afford a fucking 2tb drive.

Right now, as we speak, there are barely as many nodes as there are online players for ARK, a stupid game full of bugs, unoptimized, for which you need at least 16gb of RAM and at least a 3060  or anything with upwards of 8 gb of VRAM and at least a 2 TB SSD, because the way the update works from the stupid team that created this game you basically re-download every single damn file every time.

So if the Bitcoin community can't afford the same hardware as some guys who like to ride dinos to battles in some game, how in the name of god are we going to crush the banks, re-write the economic system and all the other bla bla bla...
Wait, revolution cancelled; a 2 TB SSD that would be enough for 4 years of blocks is $200....the banks won!!!!

Join the lukechaincoin, they made their blocks 30% the size, or even better, let's have blocks only 1 kb in size, so even after 100 years you can run your one tx per block chain on a 166 mmx with a 2.3 gb hdd.

Stompix, are you a gamer?
I felt good seeing someone on here talks about ARK, the  game sucks big time lol, I would rather spend few hours on Ark Riders

I don't understand why people are complaining, it's not as if Bitcoin Node requires gen 5 NVME SSD only, Nodes can be run on normal SSD drives that we are used to, they are even cheaper than NVMe.

A gen 3 NVMe would do just fine, even those laptops small SSDs.
Buying them is a one time thing, you aren't buying again in 2027, I just got me 4TB not too long and I don't care.

If you don't have SSD you can still use HDD too.
Still have a WD red plus from 3 years ago, 2TB still kicking ass.

[leea-1334]

==
$80 a year would be 4 ?? Maß 
Sorry, but if we can't find 50k -100k people who are willing to spend the price of 4 beers at Oktoberfest or the price of GTA 6 to challenge the old system, then the new system has failed!

Oh, did I mention that if we were to migrate to LN it will take 30 years for eveyone in the world to open a channel?

A small amount for sure, even for me but because of lets say sig earnings,,, this is the price of 2 weeks sig. The new system indeed fails. But I would like it to still be small enough that more people in other parts of the world who can't buy GTA 6 to also participate. Node locations also a part of decentralization right? I think at this rate it will be very possible. 2 halvings and 50% decrease in cost?

[ABCbits]

10 MB*144 blocks*365 days is about 525 GB per year (assuming blocks are always full). That would cost around $50–80 per year using SSD storage at today’s prices.

Bandwidth: about 1.4 GB per day to download, and around 350 GB a month to upload (assuming 8 peers on average).

The estimation of upload size per month isn't realistic, since it assume your node always receive new blocks before all 8 of connected peers. In reality, some of the peers probably already receive the new block before your node.

Good, so finally we have some numbers!
I'll ignore the metered connection because it's 2026...like f shit I have unlimited mobile data for 15 euros, common!

Out of curiosity, is it truly unlimited or have certain limitation? For example, after 10GB in a month, the maximum speed lowered to 10%.

[stompix]

Out of curiosity, is it truly unlimited or have certain limitation? For example, after 10GB in a month, the maximum speed lowered to 10%.

25G at 5G, then I have no idea cause I never saw a difference, nor do I actually know if I have passed the cap at all these months.
I have wi-fi almost everywhere, so I'm not really using data other than trips.
And the local home is 200 Mbps, no limit.

If you actually do save money, its only when there's congestion and only benefits those who do a lot of transactions.

Aka actual users, thank you for admitting this! Users! The people that had to pay those 30-40$ you cry over a year in a single fee when the spam wave hit.

Stompix, are you a gamer?
I felt good seeing someone on here talks about ARK, the  game sucks big time lol,

ARK is great because it sucks so bad!
This is the beauty of that game, the fact that you get stuck in a tree you die to a compy and your corpse is meshed 2 levels below the ground so you lose all your items..
Perfection!  

I don't understand why people are complaining, it's not as if Bitcoin Node requires gen 5 NVME SSD only, Nodes can be run on normal SSD drives that we are used to, they are even cheaper than NVMe.

A gen 3 NVMe would do just fine, even those laptops small SSDs.
Buying them is a one time thing, you aren't buying again in 2027, I just got me 4TB not too long and I don't care.

But how much would that cost on Sentinel Island? Have you thought of it?  

Node locations also a part of decentralization right?

You're not going to like this:
https://newhedge.io/bitcoin/node-map

[GeorgeJohn]

Bitcoin decentralization depends on people been able to verify transactions and enforce networks rules independently. Imagine a future where electricity bills, internet expenses keep increasing. If only wealthy individuals, large companies could afford to run nodes, would Bitcoin still be decentralized as it is today?

It seems that you're not aware that some corporate run their own private node but yet bitcoin is still operating in a decentralized form..I want you to understand that private companies can run their own nodes but it does not break Bitcoin for not be in decentralization method..most us really need to understand bitcoin, the way we looked bitcoin and it's developed technology, it's not how bitcoin works...

[Hamza2424]

Bitcoin decentralization depends on people been able to verify transactions and enforce networks rules independently. Imagine a future where electricity bills, internet expenses keep increasing. If only wealthy individuals, large companies could afford to run nodes, would Bitcoin still be decentralized as it is today?

Electricity can never be expensive, bro. You know there are solar panels, haha. These days, people are using them, and a big chunk of electricity is not being used by ordinary people anymore. The state is either selling it or using it for foreign or defence interests, which is good unless they are doing some kind of corruption, haha.

Anyway, the internet can be a problem. I just hope we get our own one without relying on these centralized authorities, but there is no way for us unless we own Elon's satellites, haha. Because the problem is not about the internet being expensive. It's about what if we don't get access to the internet for a whole month. In my country, they stopped internet services completely for at least a week once in 2024 or 2023. I don't remember exactly, but that was so stressful for many of us.

[atookz]

10 MB*144 blocks*365 days is about 525 GB per year (assuming blocks are always full). That would cost around $50–80 per year using SSD storage at today’s prices.

That's great to have some numbers in this thread now, but I believe the 10 vMB is low.

vMB isn't the same as actual MB. Real blocks are larger than their vsize, due to the discount on witness data. Full blocks are around 1.5-2 MB per vMB at the moment, so 10 vMB would be about 15-20 MB per block. This is closer to 800 GB-1 TB/year than 525 GB/year. It’s more than the $80 figure, but still affordable for most people.

The estimation of upload size per month isn't realistic, since it assume your node always receive new blocks before all 8 of connected peers. In reality, some of the peers probably already receive the new block before your node.

I agree with this. Compact blocks cut block relay a lot but transaction relay is not reduced so I would use 350 GB/month as an upperbound. It also relies on whether or not your node is accessible in any way at all.

They are already as low as they can be (1-3 sat/vb) as it is and bigger blocks won't really help bring them down much further. If you actually do save money, its only when there's congestion and only benefits those who do a lot of transactions.

Today, agreed: the fees are ~1-2 sat/vB, only about 0.6% of miner revenue. That won't change with bigger blocks, but longer-term, security will depend more on fees and block size could be important. I don't have a good answer on that, but I believe it belongs in the same trade-off.

[tbct_mt2]

Bitcoin decentralization depends on people been able to verify transactions and enforce networks rules independently. Imagine a future where electricity bills, internet expenses keep increasing. If only wealthy individuals, large companies could afford to run nodes, would Bitcoin still be decentralized as it is today?

It's not too expensive to run a Bitcoin full node, you only don't know how to run a Bitcoin full node with cheap cost. If you don't want to use SPV wallets such as Electrum wallet, you can still use Bitcoin Core and run your Bitcoin prune node with very flexible data storage depends on your setup.

[GUIDE] The simplest Full Node guide ever.
[Guide] How to run a Bitcoin Core full node for under 50 bucks!
[Guide] FULL NODE OpenSUSE 15.3: bitcoind + electrs + c-lightning + RTL

[HustleZ]

Bitcoin decentralization depends on people been able to verify transactions and enforce networks rules independently. Imagine a future where electricity bills, internet expenses keep increasing. If only wealthy individuals, large companies could afford to run nodes, would Bitcoin still be decentralized as it is today?

I think with the progressing technology it would become easier to run a node, and imo that time could also come where everyone is running a node and it becomes completely normal not something like today that is only done by a fraction of a people. These are all imaginary talks and hypothetical things, realistically looking st the trend it may become very easy to run a node

[ABCbits]

Node locations also a part of decentralization right?

You're not going to like this:
https://newhedge.io/bitcoin/node-map

That website show limited data, unless you paid their Pro subscription. I recommend to check https://btcnodes.io/nodes/live-map/ or https://btcnodes.io/nodes/ (scroll down, select "CC" button and click "All (95)") instead.

Interestingly, some country that considered as developed country or very rich (such as Japan or Qatar) have much less node compared with US.

The estimation of upload size per month isn't realistic, since it assume your node always receive new blocks before all 8 of connected peers. In reality, some of the peers probably already receive the new block before your node.

I agree with this. Compact blocks cut block relay a lot but transaction relay is not reduced so I would use 350 GB/month as an upperbound. It also relies on whether or not your node is accessible in any way at all.

I get the point, although on transaction relay, some peers may already have the TX. But in practice, you could just set maxuploadtarget on Bitcoin Core based on cap of your internet connection, rather than estimate how much GB used.

[leea-1334]

Node locations also a part of decentralization right?

You're not going to like this:
https://newhedge.io/bitcoin/node-map

Ya I thought so Many many years ago,,, a few of us thought it might be a good idea to start volunteer our computers for nodes, but even then the cost seemed too high and we are better off than many of our people, already too expensive at the time.

Still nice to see my country make it almost 1% but we are the richest in our region  

Interestingly, some country that considered as developed country or very rich (such as Japan or Qatar) have much less node compared with US.

Population explains this I guess. Japan has a big pop but majority is too old or too young, is my guess. Qatar too small,,, most of millions of people there are poor workers.

[Aanuoluwatofunmi]

Bitcoin decentralization depends on people been able to verify transactions and enforce networks rules independently. Imagine a future where electricity bills, internet expenses keep increasing. If only wealthy individuals, large companies could afford to run nodes, would Bitcoin still be decentralized as it is today?

It seems that you're not aware that some corporate run their own private node but yet bitcoin is still operating in a decentralized form..I want you to understand that private companies can run their own nodes but it does not break Bitcoin for not be in decentralization method..most us really need to understand bitcoin, the way we looked bitcoin and it's developed technology, it's not how bitcoin works...

You make the points and stress on them emphatically, no any institutional adoption that stand a threat to Bitcoin decentralization, government or any other centralized platform can decided to invest in Bitcoin or even be a giant holder, this does not change the market or anything concerning its network protocol, bitcoin is far ahead all this we may think could serve as a threat over it network.