You are viewing a plain text version of this content. The canonical link for it is here.

Posted to notifications@iotdb.apache.org by "xieqijun (Jira)" <ji...@apache.org> on 2021/11/30 15:40:00 UTC

[jira] [Commented] (IOTDB-2075) Accelerate the process of insertTablets by using thread pool

    [ https://issues.apache.org/jira/browse/IOTDB-2075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17451194#comment-17451194 ] 

xieqijun commented on IOTDB-2075:
---------------------------------

test result:

tablet count	use serial	use thread pool
100	169	166
1000	378	308
5000	1044	918
10000	1375	1497
50000	7648	7172

Because the speed of insertTable is very fast, about 0.15ms, there is no acceleration effect. In order to verify this conjecture,
insertTable sleep 10 ms

tablet count	use serial	use thread pool
100	1206	216
1000	12310	1167
5000	59562	5177


test device:
MacBook Pro (16-inch, 2019)
2.6 GHz 6-core processor Core i7
16 GB 2667 MHz DDR4

> Accelerate the process of insertTablets by using thread pool
> ------------------------------------------------------------
>
>                 Key: IOTDB-2075
>                 URL: https://issues.apache.org/jira/browse/IOTDB-2075
>             Project: Apache IoTDB
>          Issue Type: Improvement
>            Reporter: xieqijun
>            Priority: Major
>              Labels: pull-request-available
>
> Currently, insertTablets uses the serial method to execute the insertTable. If the thread pool method is used to execute multiple insertTablet in parallel, the efficiency should be improved to some extent.



--
This message was sent by Atlassian Jira
(v8.20.1#820001)