Hi, I’m new to Kopia but I love it so far. I’m wondering if I can read some details about how the “hashed chunks” process works. My repository has a mix of large and small files and I want to know for instance:
- If I change or delete a small file, will I have to re-upload a part of a large file too?
- If I have a tiny file, say a README.md file that is only a few kb, will that be its own chunk? Or do chunks include many files?
Just want to know some best practices so I can avoid taxing my country’s slow internet speeds. Thanks!
Chunks (“contents” as they are known in Kopia) are typically between 1-8MB. For smaller files, chunk is the whole file, for larger files, say 100MB it will be split into chunks, either on fixed boundary (say 25 x 4MB chunks if you configure FIXED-4M splitter) or dynamically based on rolling hash (which is the default), so chunks will be approximately 4MB.
This scheme is chosen so that when you change a very large file (say video file) by only updating metadata, typically only 1-2 contents need to be uploaded, but not the whole thing.
What may be confusing is that “contents” are packed into “blobs” (20MB+) to avoid having to deal with millions of small files in the repository.
Thanks! I’ll probably be back with more questions as I experiment with the program, but that answers my question about hashing!