Transaction Bandwidth for Bitcoin Nodes Could Drop by 75%, If New Erlay Proposal is Accepted

• Erlay proposal could reduce the bandwidth necessary for transactions.
• With the new protocol, the Bitcoin network could improve security with more connections between nodes.

The size of the nodes on Bitcoin network has been a major subject of debate, but there is a new protocol that has been brought to the table called Erlay. Erlay was proposed as a joint effort between Gleb Naumenko of the University of British Columbia and two Bitcoin developers – Greg Maxwell and Pieter Wuille. In this protocol, the “transaction bandwidth” of Bitcoin nodes would see a 75% reduction.

All of the nodes on the Bitcoin network link together, and each transaction is ultimately broadcast, as it is processed through the network. Erlay would alter the way that the transactions are executed. For a little more clarity, an email from a Bitcoin developer stated,

“The main idea is that instead of announcing every transaction to every peer, announcements are only sent directly over a small number of connections (only 8 outgoing ones). Further relay is achieved by periodically running a set reconciliation protocol over every connection between the sets of withheld announcements in both directions.”

According to developer Naumenko, the implementation of the protocol “save half of the bandwidth” of each node, which means that the connectivity will basically be increased for free and there will be a reduced risk of timing attack. Even with an increase to 32 for the outbound peer count, the current protocol is reduced by 75%. With less bandwidth, there are more connections to hold together the notes.

A good portion of the security of the Bitcoin network has at least a little bit to do with the connections made. By implementing this protocol, the increased number of connections will end up creating more security for the nodes, and that makes that the network is further protected against a network attack.

Naumenko decided to bring up the Eclipse attack, which involved a node that was

“isolated from the longest chain.”

The developer added,

“In this case, an attacker, for example, can make a target node believe that they paid that target node, without actually submitting transactions to the longest chain.”

A research paper from 2015 describes the issues in a lot more detail, which can be found here.

Considering how important the protocol appears to be to the security of Bitcoin, will Bitcoin Core use it as well? Naumenko spoke with CoinDesk and explained that a conversation with Bitcoin Core contributors was “generally positive.” However, it looks like Bitcoin Core needs to see “more experiments” before they are ready to add it.

As a general rule of thumb, Bitcoin Core doesn’t generally add new technology until the majority of active contributors and the wider ecosystem agree on it. The developer explained that there is been “positive signals” from the Bitcoin community so far, which tells the team that the continued work on the exchange is worth it. With this ongoing progress, it is likely that there will be a major release including the protocol.