What happens if blocks are generated faster than they can be downloaded?

MoonShadow

legendary

Activity: 1708

Merit: 1007

Quote from: franky1 on March 12, 2013, 10:11:33 PM

but with that said my initial reply about worrying about this issues when dialup is nearly obsolete makes the point of this moot.

Not quite. Excluding places in the world where dialup remains the most cost effective method of Internet access, there remain methods of access that are metered or otherwise bandwidth constrained that much of the population of the rest of the world depend upon for general Internet access. In these cases, a full client is unwise and a light client is better suited. For example, my android phone can handle a full client, but I pay overage charges for excess 3G bandwidth, so I use BitcoinSpinner instead of a full client.

Nicolai

newbie

Activity: 39

Merit: 0

Quote from: Lazer on March 12, 2013, 09:05:53 PM

Quote from: Nicolai on March 12, 2013, 09:04:13 PM

Max block size is 1 mb

So this should not happen (as long as the max blocksize isn't changed)

But theoretically, it can happen given enough transactions.

No. Not as long as the block size can't be > 1 mb.

Quote from: knightmb on March 12, 2013, 09:27:56 PM

Maybe, but that is not counting the CPU time and disk I/O time. I know if I haven't run bitcoin for a while and it's a few weeks behind it takes all day just to catch up.

I believe even my Nokia 3310 can download and verify 1 mb transaction within 10 minutes (exaggeration furthers understanding).

Catching up many blocks takes time, yes, but in theoretically you 'just' need to download+validate one block within 10 minutes and you will be able to catch up.

Quote from: MysteryMiner on March 12, 2013, 10:21:10 PM

If the block size remains unchanged, in far future there will be high transaction costs so no Satoshi Dick style transaction flood will take place. 1M in every 10 minutes are not that much. JewTube videos take at least 10 times as much traffic as Bitcoin under worst conditions.

See this mail from Gavin (sent 6 hours ago):

Quote from: justusranvier on March 12, 2013, 09:07:48 PM

A miner who publishes a block that propagates slowly risks an orphaned transaction. If another mines a block at about the same time, and if the second block propagates more quickly through the network, other miners will start working from the faster block before they see the slow block.

MysteryMiner

legendary

Activity: 1470

Merit: 1029

Death to enemies!

If the block size remains unchanged, in far future there will be high transaction costs so no Satoshi Dick style transaction flood will take place. 1M in every 10 minutes are not that much. JewTube videos take at least 10 times as much traffic as Bitcoin under worst conditions.

franky1

legendary

Activity: 4214

Merit: 4458

Quote from: Snowfire on March 12, 2013, 09:51:39 PM

You cannot fit an infinite amount of data into 1Mb.
So what is the actual answer to the question?

a BLOCK is not the whole BLOACKCHAIN(infinite) it is just the number of transactions every 10 minutes. which in dialup would download faster then it would be created /solved.

the only issue you would have is downloading the whole blockchain.. the string of blocks. as there are well over 200,000 of them right now.

so on dial up it would take a while, but it would happen.

but with that said my initial reply about worrying about this issues when dialup is nearly obsolete makes the point of this moot.

justusranvier

legendary

Activity: 1400

Merit: 1009

Quote from: Snowfire on March 12, 2013, 09:51:39 PM

So what is the actual answer to the question?

https://en.bitcoin.it/wiki/Scalability#Network

Quote

Let's assume an average rate of 2000tps, so just VISA. Transactions vary in size from about 0.2 kilobytes to over 1 kilobyte, but it's averaging half a kilobyte today.

That means that you need to keep up with around 8 megabits/second of transaction data (2000tps * 512 bytes) / 1024 bytes in a kilobyte / 1024 kilobytes in a megabyte = 0.97 megabytes per second * 8 = 7.8 megabits/second.

This sort of bandwidth is already common for even residential connections today, and is certainly at the low end of what colocation providers would expect to provide you with.

When blocks are solved, the current protocol will send the transactions again, even if a peer has already seen it at broadcast time. Fixing this to make blocks just list of hashes would resolve the issue and make the bandwidth needed for block broadcast negligable. So whilst this optimization isn't fully implemented today, we do not consider block transmission bandwidth here.

Snowfire

full member

Activity: 122

Merit: 100

You cannot fit an infinite amount of data into 1Mb. If the system continues to grow, and there are more and more transactions, something has to give eventually. My experience is that, from where I am, recent blocks are taking about 60 seconds to download and process. So we are still about a factor of 10 away from meeting the problem. But a traffic increase of that order of magnitude might be only one or two years down the road. No one expects global internet speed to increase by that factor in the next couple of years; it would require massive rollout of technologies like direct fiber-optic connection to the computer, something largely unrealized. So I do not think the question can be rationally evaded on the presumption that internet speed will automatically ramp up to meet any need, forever.

So what is the actual answer to the question?

knightmb

sr. member

Activity: 308

Merit: 256

Quote from: Nicolai on March 12, 2013, 09:04:13 PM

Max block size is 1 mb

So this should not happen (as long as the max blocksize isn't changed)

Maybe, but that is not counting the CPU time and disk I/O time. I know if I haven't run bitcoin for a while and it's a few weeks behind it takes all day just to catch up.

justusranvier

legendary

Activity: 1400

Merit: 1009

A miner who publishes a block that propagates slowly risks an orphaned transaction. If another mines a block at about the same time, and if the second block propagates more quickly through the network, other miners will start working from the faster block before they see the slow block.

franky1

legendary

Activity: 4214

Merit: 4458

meanwhile in the 1990's ........

only joking even dialup(56k modem) can download 1mb in under 10 minutes.

so worrying about this just shows that you need to sit down for atleast 20 seconds and think about internet speeds in the future

this is why this thread topic is a moot point to make.

Lazer

member

Activity: 71

Merit: 10

Quote from: Nicolai on March 12, 2013, 09:04:13 PM

Max block size is 1 mb

So this should not happen (as long as the max blocksize isn't changed)

But theoretically, it can happen given enough transactions.

Nicolai

newbie

Activity: 39

Merit: 0

Max block size is 1 mb

So this should not happen (as long as the max blocksize isn't changed)

Snowfire

full member

Activity: 122

Merit: 100

This may be a concern more future than present, but it would seem to be possible for blocks in a system of increasing size eventually to contain so much transaction data that many nodes can no longer download them as fast as they are created (i.e download time starts to exceed 10 min/block.) Whereas the system at present seems to depend on most nodes' being synchronized and aware of the entire block chain, this synchronization would no longer be possible under such circumstances, and that this could have adverse consequences. Nodes having the least feed bandwidth would be expected to encounter the problem first. Any thoughts?

Topic: What happens if blocks are generated faster than they can be downloaded? (Read 1361 times)