I’ve read that you should consider storing blobs over 1 MB in the file system, as opposed to couchbase. And that couchbase won’t store documents larger than 20 MB.
These documents haven’t mentioned large json files. I have some large json files between 1 and 4 MB in size. Do those same rules apply to json files in addition to binary files?
CB is a memory centric architecture so it best to under stand your access patterns.
Does all your data need to be in memory or can some be fetch from disk? What percentage? i.e. your working set.
The benefit with storing things in CB as JSON is that you can leverage things like Full text search , SQL query and more.
If you are storing large Images or video its best to put that data in a CDN closer to your end users and store the URL of that in CB as JSON.
The data needs to be stored server side. Not all of it needs to be in memory. It sounds like it makes more sense to cache server side what we need in memory. Thanks!