Janoš Guljaš
Janoš Guljaš
@jmozah Of course. I deliberately did not delete the branch.
Thanks @jmozah. This PR has nothing to do with the Forky (fcds) it replaces leveldb completely. I thought that the goal would be to replace fcds.Store, at least to measure...
> What do you suggest, i just use badger for fcds and benchmark against the forky branch? is that enough? My impression was that fcds benchmarks will be done first...
> Why?. They have a bug in the new version. it was reported only few days back and they are fixing it. There is even a patch for that. Unil...
Thanks. We should watch for the remaining issue.
I would consider this a very reasonable requirement, even very important.
I see, different prefixes for archived keys and regular ones in StateStore is the way to go. There already is the Iterate method with prefix and moving data from one...
We should see what kind of queries there would be. All cheques and balances (live and archived) can be under a single key prefix, but that they contain a field...
I believe that this is mostly suggested by @zelig, if I am not mistaken. Maybe he could elaborate more.
Can you send instructions how you managed to run benchmarks that you referenced in the description on fcds-tenage-mutants branch? These benchmarks are not existing there, and a lot of testing...