JayT106

In Feb, the Cronos network (crypto.com) just switched to using the rocksdb as the backend storage to solve the network sync delay issue (observed the node has overloaded disk I/O...

But it might be okay if the consensus storage does not support the rocksdb due to the main bottleneck is being on the SDK side.

Benchmark these metrics might need to consider the mempool conditions, like the mempool loading and the mempool size. Other metrics like - tx throughput - network traffics Also good to...

> I see the topic of TX throughput come up from time-to-time and I'm not totally sure I understand it concretely. Is this throughput from when a Tx enters the...

@creachadair thanks for the feedback, and I totally agree with you. If the indexer performs well (say within a sec), the synchronize call in consensus won't be an issue. We...

@creachadair do you think adding the genesis init height check will be reasonable? So the app ver check will only be executed when 1. the state height of the node...

https://github.com/cosmos/cosmos-sdk/issues/10791

Looks like there is an error happening during the redelegation event, https://github.com/cosmos/cosmos-sdk/blob/3458f64f0aeb68796e56dd334c04b5e066ebf089/x/staking/keeper/val_state_change.go#L82 possibly due to this error, https://github.com/cosmos/cosmos-sdk/blob/3458f64f0aeb68796e56dd334c04b5e066ebf089/x/staking/keeper/delegation.go#L886 It is a silent fail. if that possible for some operation issue...

I see, thanks for the context. So the redelegation at index 2 actually combined 3 operations in the end block event. And I agree with you, the last 2 events...

The abci++ indeed cannot solve the `low tx fee tier squeezed` issue. For the `block space reservation` feature, the Tendermint dev team may not want to implement this. The different...