Replies: 1 comment
-
|
Hi, This would be quite a large design change because we do need to compact to be able to make room in l0 (the first stage where data is first written). Perhaps the following workflow could work for you:
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hello,
First of all, I must say I really enjoy using ZeroFS, thank you!
I would like to back up my local machine daily (~400 GB of files, ranging from a few kilobytes up to 10 GB) as quickly as possible to a 9p mount on Hetzner Object Storage. My internet connection is limited to 500/200 Mbps, and I’m using rclone.
On the initial run, I noticed the following log message:
This suggests that uploads cannot keep up, which is expected; I can see that the upload bandwidth is fully saturated.
Another potential slowdown is in-flight compaction. When it is triggered, it consumes upload bandwidth and slows down the overall process. I've lowered
lsm->max_concurrent_compactionsto1, which helps a bit, but it would be more efficient if compaction could be paused (possibly over RPC?) or disabled via configuration.My ideal backup workflow would be:
Beta Was this translation helpful? Give feedback.
All reactions