Piotr Mikulski

It would be nice how have some script, that automatically detects flunky tests. We currently have more than 14 flunky tests in `integration-tests` alone: ``` 2021-12-09T07:06:40.705386Z [32m INFO nayduck::cargo_test: cmd=["cargo",...

@Rishabhraghwendra18 More work is required. I have a script, which prints all places where this issue happens.

@Rishabhraghwendra18 I started working on fixing it for `near-network`. https://github.com/near/nearcore/pull/5834 The issue is a bit complicated, as some other code changes / refactoring are required. I don't think it's a...

@Longarithm We can try to see if it makes difference to try to adjust cache sizes per column. Like in https://github.com/near/nearcore/pull/5212

> have a lot of threads. Select `perf_cliff` thread at the top and `Flame Graph` view at the bottom to get the most amusing result. It appears that int the...

> retrieve_raw_bytes I see that we have a cache in front of the database. This explains the spikes in performance. If you want to debug the issue, please try commenting...

I created a benchmark https://github.com/near/nearcore/pull/5312 It may be useful to try different database configuration.

The database is becoming slower and slower as number of keys grows ``` test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured Running unittests (/home/pmnoxx/repos/nearcore.master/target/release/deps/ibf-b2f9e4ff4ef48925) running 1 test...

I'm a big favor of `clippy`. I detected many potential issues ahead of time. It also has many useful suggestions on how to write a cleaner, more readable code.I think...