Twitter record size

Question

Hi!

Sorry for maybe stupid question;)

I am trying to model records like if they were modeled for twitter. It is interesting for me how big or small they are? Is it normal that they can be bigger than 512 bytes per record?

And what methods do you know for optimization records within json documents? I mean especial for json;) Maybe some links?

Thanks;)

Answer 1

To analyze the worst case: If you store data in utf8, it's conceivable that having all 3-byte characters means 480 bytes.

A utf8 may even be 4 bytes (4*160=640). I'm not sure this is an edge-case worth protecting against.