Duplicati VS rdedup

Petabaytlarca veri emanet edilen şu kodun %85'i C# https://github.com/duplicati/duplicati

For my parents' computer, I'm using the canary version (which should be the alpha version I think?) of Duplicati since years.

It hasn't had a release in a little while but work is still occurring. https://github.com/duplicati/duplicati/actions.

Hey have a read at : https://github.com/duplicati/duplicati/issues/4041

I just came across this while looking through their issues to see if anyone else had reported the firefox issue i'm running into. I'm starting to have serious reservations.

I like duplicacy because of the way it keeps the chunks in the file system, without a special database. This makes it scale up really well no matter how many backups you have (you can even have multiple computers saved). It's kind of beyond weird how you select what you want to backup with the symlinks (using the command line version), looks more like what one would make for himself in a weekend (not that I'm complaining about free software!) but it's been without bugs for me and extremely efficient. In contrast duplicati has a perfect interface, it's well maintained and everything but bogs down in any large backup, has stories about people recovering for weeks for a very few local TBs and I've experienced for myself this, granted in the python that is checking the sha256 checksums of the backups but it makes it slower many times (possibly hundreds of times), nobody checked from 2013 to 2021 (or did it on tiny datasets like 1GB or was content to wait for weeks even on something small-ish)?

I'm not really doing much about it anymore, but I have somewhat similar project: https://github.com/dpc/rdedup

I'm the author of https://github.com/dpc/rdedup . Threads is your answer. Make a large threadpool, process IO in it, use channels to submit work and responses if need.

rdedup (rust) - https://github.com/dpc/rdedup