Persistance of the SputnikVM execution
SputnikVM is running based on per-transaction, it means, starting 'sputnikVM' for every transaction by accepting transaction, header parameter and freeing at end of execution of transaction. It would require some resources (time) to do all these. It might affect the performance of the nodes.
Instead of above approach, can we have the feature of initiating the sputnikVM while starting the node using the same instance to execute the transactions which will save a lots of time.
As you can see in the implementation of EVM by microsoft They are creating an instance and using the same instance to run the transactions.
This feature will helps us to read config at start up. Look at #370
I'm not quite sure what you mean, but we're planing to do batched processing some time next year, if you mean that