Video and Photo Data Modeling

(Michael McKenzie) #1


Has anyone done any modeling on how to include and/or attach photo and video data? I am just curious to see how other have approached including that kind of information in their graph database.

Any links or blog posts would be helpful too.


(Szenyo) #2

Hm, interesting question. Do you refer to the metadata of these artifacts, or the pictures,videos itself?
For the second, I would store s3 links.

(Stefan Armbruster) #3

Don't add large chunks of blob data to the graph as property values. This is IMO one of the few anti-use-cases for graph databases.
Instead store the binary information somewhere e.g. AWS S3 and habe a node representing that photo/video. That node contains the S3 URL as reference and potentially some meta-data as well (size, qualitiy, codec). Using relationships you can then amend stuff like tagging, comments, ownership, permissions, ....

(Michael McKenzie) #4

Yeah I think it would obviously make sense to store a links to the photo/video on a node and some meta data to help identify the video/photo. So that make a lot of sense. Thanks!