In 2.0, Kylin has full HA !! In 1.X has close to HA - it has HA to run your queries - it doesn't have HA to calculate cubes
Check this <http://kylin.apache.org/docs15/install/kylin_cluster.html>: you can have 'n' queries nodes (to solve node) and use a Load Balancer to distribute the load (MailList <http://mail-archives.apache.org/mod_mbox/kylin-user/201701.mbox/%3CF67A65A186263E4F9EF49BEAEEB50CFE2EAD5D9B03%40ME-MEL-EX1.migrationexpert.com%3E> ) About the job node, You need it to create new cubes and job scheduler, in KYLIN-2006 <https://issues.apache.org/jira/browse/KYLIN-2006> v2.0.0 has HA Either way, with 1.6 if job node fail they auto-resume pending task at failure point (without data lost) (MailList <http://apache-kylin.74782.x6.nabble.com/Kylin-Job-Server-is-single-point-td6336.html#a6342> ) 2017-02-10 4:07 GMT+01:00 xingpeng <xpxingp...@126.com>: > HI, > > Does kylin support the high-availability mode that uses multiple job > instance: one active job instance and several backups in case it fails？ > > I want to enable kylin job instance high availability.What should I do > > Thanks > > -- > View this message in context: http://apache-kylin.74782.x6. > nabble.com/Does-kylin-support-the-high-availability-mode- > that-uses-multiple-job-instance-one-active-job-instanc-tp7153.html > Sent from the Apache Kylin mailing list archive at Nabble.com. >
