Disk usage: tags vs fields


Is there a significant disk usage difference between keeping some data in an array or in an individual field? The data won't be frequently queried but sometimes may act as a filter. So I'm not concerned about search speed in this case, only if I can spare a couple of GBs per X million documents.

Thank you!

I'm not entirely sure I understand. Are you asking the difference between:

  "tags": ["foo", "bar"]


  "tag1": "foo",
  "tag2": "bar"


Yes. Currently, I have about 5 values which aren't that important (e.g., I don't use scoring). With around documents would I spare disk space if those were in an array (like the tags example)?

the tags example will be more efficient


This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.