Package: rdfind Version: 1.3.5-1 Severity: normal rdfind bareally uses single core on my 8 core machine.
I do have some highly duplicate files or similar size, about 500GB is size, with few million files. Using async io, or multiple threads to do things like tree traversal, stat, open/read/close, and for hashing, would speed it up many many times, as the IO latency is a bottlneck often. Not all parts can be easilly parallelized, and some additional synchronization and granularity level needs to be decide, but still I think it might be worth it. Regards, Witold -- System Information: Debian Release: buster/sid APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 4.16.0-2-amd64 (SMP w/12 CPU cores) Locale: LANG=pl_PL.utf8, LC_CTYPE=pl_PL.utf8 (charmap=UTF-8), LANGUAGE=pl_PL.utf8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages rdfind depends on: ii libc6 2.27-3 ii libgcc1 1:8.1.0-5 ii libnettle6 3.4-1 ii libstdc++6 8.1.0-5 rdfind recommends no packages. rdfind suggests no packages. -- no debconf information
