[
https://issues.apache.org/jira/browse/IOTDB-4581?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17616877#comment-17616877
]
Haiming Zhu commented on IOTDB-4581:
------------------------------------
This is not a bug, but maybe we should find a better way to compute compression
ratio.
This phenomenon is caused by tsfile contains few series whose data can be
compressed with high compression ratio. Take
avg_series_point_num_threshold=5000 as an example:
*Scenario 1.* Small TsFile with high compression ratio, It's
*Scenario 2.* Small TsFile with low compression ratio
> Compression Ratio compute error
> -------------------------------
>
> Key: IOTDB-4581
> URL: https://issues.apache.org/jira/browse/IOTDB-4581
> Project: Apache IoTDB
> Issue Type: Bug
> Reporter: Jialin Qiao
> Assignee: Haiming Zhu
> Priority: Minor
> Attachments: high ratio data.png, high ratio log.png, low ratio
> data.png, low ratio log.png
>
>
> We try to read a tsfile and write it to IoTDB with different config but in
> same version. If we set the avg_series_point_num_threshold to 5000, 10000 and
> 100000, IoTDB computes compression rate as 100, 10 and 3. But the file size
> in the disk is the same. The compression ratio must be computed wrong.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
