Progress on the network

On the background

Progress on the bitcoin network means not only more, smaller, faster or cheaper transactions, but also improving the health and robustness of the network itself. Steady progress is also being made here, although this probably goes unnoticed by many.

A more robust network

The bitcoin developers pay attention to preventing irregularities on the network and passing on transactions and blocks over the network as smoothly as possible. If bitcoin is to be a global means of payment, it is important that the network is always reliable. An occasional malfunction at a bank is already annoying, but such a malfunction should never occur in an ambitious project such as bitcoin. So far Bitcoin is doing well in that regard, but there is always room for improvement and optimization. The time spent on making the network efficient is therefore well spent.

One of the areas where progress is being made here is the forwarding of blocks over the network. It is important that this happens quickly and that a newly found block is made known to all participants in the network as quickly as possible to prevent the network from operating incorrectly or inconsistently. If this – at first sight simple – part of the network does not function properly, transactions may receive a false confirmation, for example, which will later be reversed. This would compromise the reliability of a confirmation; something that must of course be avoided. A confirmation must be final, across the entire network.

In addition, it would be disadvantageous for some miners if blocks are sent slowly over the network. For example, it could be that the miners who receive a block quickly have a consistent head start in finding the next block, even though the computational power they have is equal to the miners who receive the block later. This makes the ‘race’ to find the next block unfair. Consistency across the network and efficiency in forwarding information is therefore important.

Progress visualized

Researchers from the Decentralized Systems and Network Services Research Group (DSN) of the Karlsruher Institut f??r Technologie (KIT) have been monitoring the Bitcoin network for the past three years and have created interesting visualizations based on the collected data. The first visualization below shows the distribution of a block across the network in 2015. Each dot on the map is a node in the network.

The spread is not very fast and could lead to the aforementioned problems in the longer term. The same visualization is shown below, but for 2018. It is clear how much more efficiently the network spreads a block – a healthier network. While there are of course more nodes on the network today than in 2015, most of the progress is due to network improvements.

Progress doesn’t always mean faster

It seems logical to also make the spreading of transactions across the network increasingly efficient. This is partly true. If there are ways to make transactions smaller so that it takes less bandwidth to distribute them, then such an optimization is definitely worth it. However, improving the speed of the spread of transactions is not so obvious.

After all, a fair amount of (personal) information can be retrieved based on the distribution of transactions. It is likely that the first node to pass a transaction is the node from which the transaction originated. This would allow the IP address of a node to be linked to a transaction. This is an anonymity issue. In order to carry out such de-anonymisation, it is necessary that the ‘attacker’, i.e. the person trying to discover identities, has a reasonably large number of nodes in the network in order to be able to keep a close eye on where transactions come from. come.

Many bitcoin analysis tools, such as Chainalysis, work with similar techniques. On, for example, you can also see which node (read: which IP address) reported the transaction first to the node. However, the reliability of the information from such services depends on the number of nodes to which the service is directly connected. So it can be beneficial for user privacy to make it a bit more difficult to find out where a transaction came from in the first place.

This is achieved by applying a small delay when spreading transactions across the network. This delay makes it more difficult to determine from which node a transaction originated in the first place. When one node forwards a transaction to a single other node and both nodes only forward to subsequent nodes after a delay, it becomes difficult for subsequent nodes to determine which of the two first nodes originated the transaction.??

Compared to the distribution in 2018, which can be seen in the visualization below, transactions in 2015 were transmitted faster than today. Note that approximately the same amount of nodes is reached in both visualizations in about three seconds; especially at the beginning of the spread, the delay does its job to make it more difficult to identify the origin and thus does not slow down the network overall.

While the delay doesn’t make it impossible to trace the originating IP address, it does make it a bit more difficult. How much more difficult is unclear, and a subject that is being researched. Improving and making privacy robust is therefore not done with a single trick, but requires a convergence of different approaches.

View all visualizations on the DSN website.

Related Posts

Leave a Reply

Your email address will not be published. Required fields are marked *

© 2024 Cryptocoin