Hi Experts, I am sqooping out a Hive table with 20-30 columns and apprx. 100 millions records to Oracle. sqoop export -D mapred.job.queue.name=<yarn queue name> -Dsqoop.export.records.per.statement=100000 --connect jdbc:oracle:thin:<db server> --username xxxx --P --table employee --hcatalog-database default --hcatalog-table employee --num-mappers 10 --batch Source Hive table is unpartitioned and has 200 part files, total 1.1 GB of data. And destination oracle table has identity defined on pk column which has cache of 100000. It's taking approx 5-6 minutes to sqoop out data completely with default fetch size 1000. Any suggestion how can I improve more to get better performance. Is there anyway I can change fetch size? Thanks in advance. ... View more

Hi Experts, I want to offload data from my hive table to apache cassandra cluster and was evaluating possible options. Tables size can vary from small to large. Is any tool other than sqoop and spark can someone recommend which is being used mostly and have performance benefit as well. Thanks in advance! ... View more