Hi @timsilva,
I think most of what you are seeing is just overhead, as odrive is trying to index and track millions of objects. This sounds like a pretty massive data set you are bulk uploading, so I highly recommend trying to break up the set.
odrive needs to scan and track that entire set in real-time, both on the local and remote sides. Since odrive is full sync rather than traditional backup, the processing is heavier. Breaking up the dataset and letting the “segments” sync, before adding the next can help a lot.
Another thing that can be extremely helpful is to unsync sections that have already completed syncing to the cloud. This reduces the terrain that odrive needs to cover and relieves overhead processing.
The larger the dataset, the more work is required, and that work is constant. I expanded on this a bit in this post: Backup (one-way sync) - #11 by Tony
The odrive menu tends to get bogged down with this amount of activity, so you can also try using the CLI to help with odrive status. Check out this post for information on that:
These settings are more to compensate for the local environment. The logic tries to be adaptive, with the intention to leave some amount of bandwidth for other systems tasks (web browsing, for example). Unlimited doesn’t impose any limits, Normal tries to allow for about 10-15% network for other things. Limited tries for about 40-50%.