[
https://issues.apache.org/jira/browse/IOTDB-3668?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17559167#comment-17559167
]
Haiming Zhu commented on IOTDB-3668:
------------------------------------
The related PR is https://github.com/apache/iotdb/pull/3138
> Tiered Storage
> --------------
>
> Key: IOTDB-3668
> URL: https://issues.apache.org/jira/browse/IOTDB-3668
> Project: Apache IoTDB
> Issue Type: Task
> Components: Core/Engine
> Reporter: Haiming Zhu
> Priority: Major
>
> With the rapid development of the Internet of Things, the volume of real-time
> monitoring data generated by devices has increased rapidly. Besides, in order
> to support industrial Internet applications, enterprises have gradually
> increased their requirements for storing full historical data, which
> increases storage costs.
>
> However, the time series data application load has the characteristic of
> ’newly written data has a higher probability of being queried, historical
> data has a lower probability of being queried, and a little higher query
> delay is acceptable’. To this end, a feasible way to reduce storage costs is
> to use tiered physical storage media: group the storage medium by
> performance, price, and capacity to form a hierarchy. Hot data is stored on
> the high-performance storage medium layer and cold data is stored on the
> low-performance storage media layer. When the nature of data changes, the
> migration mechanism is used to migrate data to the appropriate storage layer.
> By optimizing the distribution of data, databases can achieve the best read
> and write performance.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
