Uploaded image for project: 'Hive'
  1. Hive
  2. HIVE-15690

Speed up WebHCat DDL Response Time by Using JDBC

    XMLWordPrintableJSON

Details

    • Hide
      This change is already implemented and heavily used at www.datarepublic.io powering our REST based product console.

      PR is raised against our Hive clone back to Apache: https://github.com/apache/hive/pull/133
      Show
      This change is already implemented and heavily used at www.datarepublic.io powering our REST based product console. PR is raised against our Hive clone back to Apache: https://github.com/apache/hive/pull/133

    Description

      WebHCat launches new hcat scripts for each DDL call which makes it unsuitable for interactive REST environments.

      This change to speed up /ddl query calls by running them over JDBC connection to Hive thrift server.

      Also being JDBC connection, this is secure and compatible with all access policies define in Hive server2. User does not have metadata visibility over other databases (which is the case in hcat mode.)

      Attachments

        Issue Links

          Activity

            People

              abbaspour Amin Abbaspour
              abbaspour Amin Abbaspour
              Votes:
              1 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

                Created:
                Updated:

                Time Tracking

                  Estimated:
                  Original Estimate - 24h
                  24h
                  Remaining:
                  Time Spent - 20m Remaining Estimate - 23h 40m
                  23h 40m
                  Logged:
                  Time Spent - 20m Remaining Estimate - 23h 40m
                  20m