[
https://issues.apache.org/jira/browse/IOTDB-749?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17127988#comment-17127988
]
Jialin Qiao commented on IOTDB-749:
-----------------------------------
Hi, we do return batches of results at present, the default fetch size is 10k.
However, the number of columns is not restricted.
This issue is for the scenario that users select 1M columns at a query, then
one batch contains 1M * 10k points and the memory may not able to cache one
batch...
Adjusting the fetch size maybe a solution, but these queries will too slow. I
prefer to reject these queries at once if needed.
> Handle select * from root OOM
> -----------------------------
>
> Key: IOTDB-749
> URL: https://issues.apache.org/jira/browse/IOTDB-749
> Project: Apache IoTDB
> Issue Type: Improvement
> Components: Core/Engine
> Reporter: Jialin Qiao
> Priority: Major
>
> In this query, select * from root
> the server will cache [the total time series * fetch size] points. This may
> cause OOM, we need to handle this.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
