Hello!
I am a consultant in digital imaging workflows, including DAM, with over 25 years experience. I am always on the lookout for good products and would love to speak with someone at the company to get more information. Please let me know how I can contact a relevant person. Could not find another way to contact someone internally.
Thank you.
Howard Goldstein
hrgoldstein@gmail.com
http://www.digitalimagingstrategies.com
I contacted you about 4 years ago. I currently have a client, a music photographer with 30 years or so of images- about 120TB on a QNAP NAS storage unit. Probably multi-millions of images - JPEG, various RAW, some video. They were running Extensis Portfolio previously on a Mac platform which is no longer supported and had performance issues. We have been trying to run an internal install of Resourcespace on an Ubuntu server and have had issues with that even with Montala support. We have about half a dozen clients, with some accessing via a VPN. The NAS and the Ubuntu server are on a 10GB network. Could we run the server version on the Ubuntu machine ( or another version of Unix) and have a catalog of multi million assets, or would we need to have multiple catalogs? If the latter, is there a way to cross search between multiple catalogs? I would love to have a discussion about this in real time if possible. Thank you in advance. Howard Goldstein hrgoldstein@gmail.com http://www.digitalimagingstrategies.com
Hello Howard, from the look of it, the number of images may be at least stretching the limits of a fully distributed system like this. The main issue is that there are several places where it’s necessary to iterate through the whole library, for example when scanning the file system or when synchronizing between devices. Things can be optimized to a certain degree, but there will be a point at which certain operations will become slower than they should.
Also, the thumbnail/preview cache stored on each of the clients will start to take up a considerable amount of space with a large number of images. Assuming roughly 250 KB per image, 10 mio. images would mean 2,5 TB of storage needed on every client. This could be mitigated by adding the option to only store the 256x256 thumbnail size locally, though.
A simple centralist system with black-box storage and a database-like query interface on the other hand is much easier to scale in terms of image count, but that would contradict our design goals and is not really an option.
In order to judge whether this still falls within a reasonable range, how many millions of assets are we roughly talking about? We are so far testing with a library of 1 million images to ensure that works decently, and a few million should also still be fine given enough RAM. We can set up a call to go through the details or to think about possible approaches.