There is actually a way to change the numbers of mappers in Pig. Pig uses a CombineFileInputFormat to merge small files into bigger map tasks. This is enabled by default and can be modified with the following parameters: For the rest what Artem said. pig.maxCombinedSplitSize – Specifies the size, in bytes, of data to be processed by a single map. Smaller files are combined untill this size is reached. pig.splitCombination – Turns combine split files on or off (set to “true” by default). ... View more

Phoenix is a library always added to HBase. There are also Phoenix Query servers which are essentially Phoenix proxies for clients who cannot access all data nodes. They can be added to a host on the host page Click on a host, press +Add: Phoenix Query Server ( more like a client than a service since Phoenix is already part of HBAse ) ... View more

Are you sure that uit works. I.e. have you restarted the Hive services after changing authentication to LDAP. I am pretty sure it should block you after. I did PAM before a couple of times and I assure you when I enable that nothing works anymore if any of the other dependencies are not setup correctly. https://community.hortonworks.com/articles/591/using-hive-with-pam-authentication.html ... View more

I wrote a little article on getting started with SQL STD Auth. I implemented it and did quite like it. https://community.hortonworks.com/content/kbentry/597/getting-started-with-sqlstdauth.html ... View more

By itself any HDFS user can use Hive. If you do not have kerberos: 1) Just add the user to Linux with useradd ( or add to your ldap ) 2) create user home directory as user hdfs do hadoop fs -mkdir /home/myuser hadoop fs-chown myuser /home/myuser And that's it you should now be able to run queries as that user. Now there is a second part for secure clusters If you have kerberos: 1) Add user to your KDC 2) kinit as the user 3) run hive Using hiveserver2 with beeline or jdbc driver: Depends on which security you have configured for hiveserver2 - None, just specify myuser as user ( -n myuser ) - PAM uses Linux so give -n myuser -p myuserlinuxpassword - LDAP uses an LDAP server lets assume its the same as your linux user - kerberos needs a kinit and specify the principal in the jdbc url Finally your user needs access to the tables: No authorization - make sure the tables you want to read are readable by that user or enable doAs=false Ranger - Add access to the table in Ranger portal SQLStdAuth - Grant access to the table using GRANT command ... View more