SUMMARY Maybe I have not the right constellation by using digikam, but on Windows it is incredible slow. I would like to put a digikam server version on the wishlist to run long time processing tasks. STEPS TO REPRODUCE My constellation: Photos on a Synology NAS. MariaDB on the same NAS. My Windows PC is connected via ethernet. In my network I run an ubuntu server. EXPECTED RESULT I can install a server version for ubuntu. Optional: I can maintain it with my Windows installation (connection via SSH and command line eg.). From the server I can start face detection and recognition, thumbnail creation, hash key generation, similar photo scanning,... Some things can run as cronjob, some as batch job.
More or less a duplicate of bug 254099. There is a wish to run digiKam on the server to run the collection scan. The question is rather where the performance problem is with you? When reading the files from the NAS? Do you use a drive path or a UNC path to access the image collection in digiKam? Maik
Furthermore, it is rather not advisable to have files and database on the same NAS. A NAS is nowhere near as powerful as a real server. Maik
(In reply to Maik Qualmann from comment #1) > More or less a duplicate of bug 254099. There is a wish to run digiKam on > the server to run the collection scan. > > The question is rather where the performance problem is with you? When > reading the files from the NAS? Do you use a drive path or a UNC path to > access the image collection in digiKam? > > Maik Great, that someone else had this idea. I tried multiple approaches to reach my photos. From technical perspective I don't see a difference but worth to mention. I added the photo directory as a network drive on windows. Then I said to digikam load this local storage. Second approach, I added the server path to digikam as network collection. No difference in speed. Which I was assume because both ways always relay on samba network sharing.
Just to give you a feeling. I started over with my MariaDB. So scanning of new entries is currently running. Started 45 minutes ago and 10% has been reached. Don't ask about face detection. I think if I run this on my whole collection, I will need 24 hours. Approx. 55K photos should be loaded.
*** Bug 435349 has been marked as a duplicate of this bug. ***