Update druid webrequest_sampled load job

  • Use time filters reducing code
  • Make TMP folder and hive table names hourly instead of daily
  • Use local execution instead of skein for spark small actions

Bug: T415359

Merge request reports

Loading