I have around 400 disk images (4-6GB each) that I have downloaded (created by a 3rd party supplier). They are all very similar Linux (raspbian) images with slightly different configurations (I do not know exactly what differences there are, but they are 1-3 base images with minor changes for each individual unit sold by our company).
This sounds to me like something that I should be able to highly compress into a single file, but I have no idea where to start for maximum compression without having to uncompress the entire file to retrieve an individual image.
Should I just be zipping all the individual .img files into one file, or should I be mounting and digging the images before compressing only the diffs (with a single base image)?
I do not have access to the list of changes performed to each file, so I would have to diff the disks in some way to find the similarities.
In case it matters, I’m running on a Windows machine, but WSL (Debian) is an option.
The images have had the tool pishrink used on them, so there is minimal unused space on each image.
