Roman Gershman comments

Results 808 comments of


                                            Roman Gershman

high CPU during slots migrations on the destination shard

what's a full slot/half slot?

chore: reduce boilerplate by removing redundant function declarations

augment review

chore: reduce boilerplate by removing redundant function declarations

@dranikpg I removed from two families if you are positive with this code removal - I will proceed with others.

Load test Dragonfly for processing pub sub commands; When the request volume reaches 40W TPS, a performance bottleneck issue occurs.

what's a 40W?

Throttle full sync replication on master

@mkaruza you recently improved journal streamer and you had some good ideas there. I would like you to start thinking about this when you finish your work on GEO search....

partial sync buffers should be bound by memory usage and not element size

@applike-ss we won't be able to address this issue in the near term. As a workaround i suggest increasing the memory reservation of your dragonfly processes. Alternatively, you can try...

Allow pipeline offloading to disk

The problem is that it's not possible to recognize when using synchronous interface because before you call `Send( )` you can not know whether it will get stuck or not....

Allow pipeline offloading to disk

I think the misconception is that I asked to accurately recognise a blocked socket and do the offloading after the fact. This is not what I am suggesting at all....

dragonfly operator namespace override

Moved to dragonfly-operator repository

Improve performance by pre-allocating vectors

1. it's a minor issue 2. `It can be more efficient (to be tested if it actually is) ` is a key question. A reproducible example with proof where we...