Hi all, I have a MongoDB with a huge set of prestored data (~20G. with text and metadata in each document). I'm using SimpleMongoReader and trying to create index. But when it needs to do reader.load_data, I ran out of RAM. Is there any workaround that I don't need to firstly load everything into memory to create index?
Hmm, I think the only solution here is not query mongo so that you aren't pulling in all 20GBs of data at once
Although creating an index over 20GB of data will use a LOT of compute/memory. I would definitely use a production vector db for this (qdrant, weaviate, pinecone)