Solving the transaction quality trilemma is imperative if we want free/negligible cost transactions without the bloat 

https://polynya.mirror.xyz/ibwxujFop29zEXhMWQrd_ujxKJvOYNQr8_hlYxk1GfE

Top 3 unsolved problems and under-researched areas in infra, IMHO: 

1) Transaction quality trilemma

...
2) Single-thread bottleneck (parallelize transactions on the same smart contract across multiple threads & machines/seqnrs) 
3) State hibernation (long-term sustainability)

Top 3 unsolved problems and under-researched areas in infra, IMHO: 

1) Transaction quality trilemma

...
2) Single-thread bottleneck (parallelize transactions on the same smart contract across multiple threads & machines/seqnrs) 
3) State hibernation (long-term sustainability) 

https://warpcast.com/polynya/0xcb430c27

Reth is exploring parallel transaction execution: https://www.paradigm.xyz/2024/04/reth-perf. Scroll published a parallel ZK-VM paper recently, and StarkNet has it on the roadmap for 2024. Monad is a standalone L1, but it's still worth mentioning in the parallel (E)VM conversation. Also adding Fuel/Eclipse to the list.

I really like the idea of DAGs for eliminating bottlenecking. In principle, you should get the advantages of shards or L2s but in a way that's entirely transparent to users.

I'm not clear what the unsolved problems are with DAGs. I imagine that achieving transaction finality is a PITA.

What do you think about the current research status of  Parallelism? Is there any worthy progress?