My apologies if this is not the correct community for my question.
I need to store time specific data in my couchbase db. More specifically I have a list of keywords and a crawler that will run every day and return quite a bit of data for the keyword. I need to keep all of the historical data. So for example, keyword "A" would have associated data for each day. The structure of the data will be consistent but the values will differ every time the crawler runs. I will need to graph the changes in the data over time. What is the best way to store the information? Should the data for each day for a given keyword be stored in separate documents or should the data from every day be stored as sub documents within a single document (perhaps with the keyword as the key)?
Why don't you do it by month of year.
This way the json document does not get to big b/c it only has 28-31 days and you has a max 12 documents per keyword per year.
To get all the data for the year you make a BulkGet([keyword_2012-07,keyword_2012-06])