[jira] [Assigned] (HIVE-22221) Llap external client - Need to reduce LlapBaseInputFormat#getSplits() footprint

["Shubham Chaurasia (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HIVE-22221?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Shubham Chaurasia reassigned HIVE-22221:
----------------------------------------


> Llap external client - Need to reduce LlapBaseInputFormat#getSplits() footprint  
> ---------------------------------------------------------------------------------
>
>                 Key: HIVE-22221
>                 URL: https://issues.apache.org/jira/browse/HIVE-22221
>             Project: Hive
>          Issue Type: Bug
>          Components: llap, UDF
>            Reporter: Shubham Chaurasia
>            Assignee: Shubham Chaurasia
>            Priority: Major
>
> While querying through llap external client, LlapBaseInputFormat#getSplits() invokes get_splits() (GenericUDTFGetSplits) udtf under the hoods.
> GenericUDTFGetSplits returns LlapInputSplit in which planBytes[] occupies around 90% of the split size.
> Depending on data size/partitions and plan,  LlapInputSplit can grow upto 1mb with planBytes[] being common to all the splits and occupying more than 850 kb. Also, it sometimes causes OOM on HS2 depending on HS2 heap size.
> This can be resolved by separating out common parts from actual splits and reassembling them at client side. 
> We can also provide an option where client can say it does not want to reassemble them and can take the control of reassembling in it's hands.
> Splits can be broken like:
> 1) schema split
> 2) plan split
> 3) actual split 1
> 4) actual split 2....and so on.
> This greatly reduces the memory(in my case from 5GB(~5000 splits) to around 15MB) on server side  and hence the data transfer. And this eliminates OOM on HS2 side.
> cc [~jdere] [~sankarh] [~thejas]



--
This message was sent by Atlassian Jira
(v8.3.4#803005)