hello Team is unnest giving any performance issues...
# general
s
hello Team is unnest giving any performance issues on filtering that values on billion records?we want to use unnest mechanisim insted creating individual column and update dse file
e
@San Kumar are you using any indexes on the values that you want to filter? That will likely make a huge difference if you’re not already, like an inverted index[1]. Partitioning can also provide huge benefits to exclude large swaths of data at the segment level during query execution. [1] https://docs.pinot.apache.org/basics/indexing/inverted-index
s
HI Evan as i understood ,we have to create index for that.and is there any reference to create index for unset function usecase.As I understood it will create a default index for this whole json .but we want on particular unset column values