Current gas limits allow you to, I think store 200 transactions. So even if we could send more, it doesn't really make sense to propagate more than for the next like two or three blocks. We are sending like 600 transactions every second currently in party, Ethereum on the mainnet because this is how much we actually have to send. But it would be really easy to send way more. It would just like increase your CPU usage and bandwidth on the network. There's a difference here between what a node can do and what a node needs to do.
In this episode, we are joined by Tomasz Drwięga, a Core Developer at Parity Technologies, to discuss the lifecycle of a transaction on the Ethereum network and how the mempool works.
We will be covering the following topics:
- What a mempool/transaction queue/transaction pool is.
- How a transaction reaches a mempool and what the mempool does with it.
- Looking at what causes the CPU increase and delays in the network.
- What happens when a transaction gets stuck.
- Gossip.
- The security properties of the mempool.
- What a network attack could look like.
Here are some additional links and ressources if you want to dig deeper.