pimlicolabs/alto

Add mempool persistence logic using Redis

Closed this issue · 5 comments

kristofgazso commented

NOTE: Lower priority

Currently we don't persist incoming UserOperations in a mempool, and send them straight to be included on-chain. This is not a big problem right now because there isn't enough volume on-chain to justify waiting for multiple UserOperations until submission for inclusion, but this is required to be compliant with the spec.

Let's create a persistent mempool using Redis with ioredis

kanthgithub commented

can this be done via rocksdb where its key-value pair?
thinking about redis vs rocksdb terms of horizontal scalability

kanthgithub commented

@mouseless-eth will be picking this up if this is still relevant change to be done
i was trying to run bundler locally with anvil

defistar commented

@mouseless-eth is this still a priority item to be picked up? i have some time this & next week and can pick this up if this is needed

mouseless0x commented

This issue is still relevant and a feature that we will most likely implement in the future, let me know if anyone would like to give it a go. Happy to discuss further in dms at t.me/mouseless