Mesh growth

[lrettig]

What are our latest estimates on mesh growth over time? Given how out of hand this has gotten in Ethereum recently (it's currently 300-400gb) - and the fact that Spacemesh will have greater throughput, and a similar VM and account model - I'm concerned that, pretty soon, you would no longer be able to run a Spacemesh full node on commodity hardware (which sort of defeats the purpose of our economic model). Of course, this also slows down syncing of new nodes.

cf. spacemeshos/research#47

[avive]

We decided that we want to build the model based on the new implemented consensus protocol changes so it is realistic. We know that the current estimate is going to be irrelevant due to protocol related changes in blocks voting, ATX sizes, ATX frequency, etc... The plan is to build a model once the big protocol changes are implemented and see if we need to further optimize before mainent based on the model results. e.g. yearly expected growth rate under moderate number of transactions.