1) Your cluster should have an application master UI, usually located at <application timeline host>:8088/cluster This will list all the running queries/applications and allow you to drill in for statistics like mappers, reducers, etc. 2) You want to use the Yarn capacity scheduler. You can create queues and allocate a % of resources to each queue. The easiest way to use the capacity scheduler is to download the Ambari view – assuming that you’re using Ambari. I’m not sure what the cloudera equivalent is.
From: mahender bigdata [mailto:mahender.bigd...@outlook.com] Sent: Sunday, February 28, 2016 2:10 PM To: user@hive.apache.org Subject: How to Query running in background in tez Hi, I have 2 queries regarding Hive Query 1. Is there a way to know which Hive Query is running in background by application ID, I would also like to know location of Log during running of the hive query in TEZ mode 2. If I'm having cluster 20 Nodes, If I submit a query, query takes entire resources of cluster. Is there way to limit the usage. As Query is taking full resources of cluster, any query which is submitted latter, will be in Pending.Is there a way to restrict to 50% -70% Cluster resource consumption for a query. Allowing other query esp. smaller query to complete. Thanks
