Re: When to use PARTITION BY HASH?

Ron <ronljohnsonjr@xxxxxxxxx> · Mon, 8 Jun 2020 08:07:04 -0500



    On 6/8/20 3:40 AM, Oleksandr Shulgin wrote:

    [snip]

    
          I've found the original commit adding this feature in
            version 11: https://git.postgresql.org/gitweb/?p=postgresql.git;a=commit;h=1aba8e651ac3e37e1d2d875842de1e0ed22a651e
          It says:
          

          "Hash partitioning is useful when you want to partition a
            growing data

            set evenly.  This can be useful to keep table sizes
            reasonable, which

            makes maintenance operations such as VACUUM faster, or to
            enable

            partition-wise join."

          
    How does hashed (meaning "randomly?) distribution of records make
    partition-wise joins more efficient?

    
    Or -- since I interpret that as having to do with "locality of data"
    -- am I misunderstanding the meaning of "partition-wise joins"?

    
    -- 

      Angular momentum makes the world go 'round.