ParPar VS MultiPar

PAR2 creation (file-based error correction): the core routine here is GF(2^16) matrix multiplication. AVX512 has the GF2P8AFFINEQB instruction, which isn't present on any GPU that I know of, but massively accelerates GF multiplication. Without this instruction, you have to resort to emulating it via bitwise operations or lookup tables, which are much slower. I actually have an OpenCL implementation, but its performance is rather meh compared to CPU

Did you tried to use parpar for creating par2 files? Supposed to be much faster. https://github.com/animetosho/ParPar

I've been developing a performance oriented PAR2 tool (used for generating file-based redundancy for error correction), but without access to a lot of hardware, I don't have that much ability to tune it to various microarchitectures. Currently there's no real benchmark setup defined/implemented (though it shouldn't be hard to come up with one), and it certainly isn't tuned for the number of cores on EPYC, but I've never really thought PAR2 benchmarks would interest that many people.

QuickPar is ancient at this stage, however the PAR2 format hasn't changed since, so the concepts are still the same. For a more modern client, I generally recommend MultiPar. I've also written a PAR2 client, ParPar (CLI, or experimental GUI frontend), but it only supports creating PAR2s, not verify/repair.

Also this implementation of par is the fastest I've seen (for creation only thought) https://github.com/animetosho/ParPar

For PAR2, I did a benchmark comparison here of every tool, which you can use to make up your mind. par2j (Multipar) and phpar2 are Windows only, but work via Wine. ParPar only supports create, so you'll need to use a different client for verify/repair.

but... depending on how much effort you want to spend fixing it I might be able to fix your current corrupt files, especially if the corrupt files are only mildly corrupt I could create a '.par2' file from mine (assuming you are on build 1436.28), then you would use MultiPar (i.e. https://github.com/Yutaka-Sawada/MultiPar/releases ) to scan your files and it will tell me how many blocks you need to fix it and then I could create recovery data with MultiPar. doing this might dramatically lower the amount of data I would have to upload as then you could use MultiPar which will fix your current corrupt files so they match mine identically.

MultiPar is the continuation for Windows https://github.com/Yutaka-Sawada/MultiPar

And this for Linux https://github.com/animetosho/par2cmdline-turbo

* MultiPar supports in-situ repair, which bypasses this problem entirely

To detect bitrot I use dynamix file integrity plugin, it compute Blake3 hash for each file, it is stored as extend attribute. You Can choose which share will be covered. For correction i then use par2 files using multipar https://hp.vector.co.jp/authors/VA021385/ https://github.com/Yutaka-Sawada/MultiPar (Windows only, par2cmdline for Linux. You could compile from source and add it to unraid if you want, i've done it for testing purpose)

And last but not less important - make regular parity files of your healthy important data, and store them somewhere. MultiPar is your best friend. Even if your data gets corrupt, with parity files you can restore it to it's former glory. Some limitations will apply, read the manual.

This is possible with MultiPar https://github.com/Yutaka-Sawada/MultiPar

More sanely, perhaps, get some M-Disc DVDs (if you can find real ones - apparently they're becoming just normal DVDs) and put one file on each, and then run https://github.com/Yutaka-Sawada/MultiPar on it to create parity files. Then store the file, the parity files, and the software you need (MultiPar) on the disc. To be safe, you probably also need to store the DVD burner you used to make it, as those are getting thin on the ground these days.

Note about par2: if one is using a windows system with a nvidia GPU this version of par2 is faster as it speaks CUDA.

For any file repairs use multipar, it's still being updated and has gpu acceleration. https://github.com/Yutaka-Sawada/MultiPar/releases