[GitHub] [iotdb] choubenson opened a new pull request, #6743: [IOTDB-3777]Avoid holding sg write lock in the whole processin of deletion

[GitBox <gi...@apache.org>]

choubenson opened a new pull request, #6743:
URL: https://github.com/apache/iotdb/pull/6743

   **Description**
   1. The write lock of sg is held during the entire deletion process (including serialized mods), resulting in "fake death" when multiple delete operations are concurrently performed.
   2. When there is a flushing memtable, no matter whether the data meets the conditions, the mods will be serialized after the memtable is successfully flushed.
   
   **Improvements**
   1. In the deletion process, in order to avoid the concurrent problem of writing and memory data deletion, it is necessary to add a write lock to the data deletion of the unsealed file, while the data deletion of the sealed file, that is, serializing mods, does not need to add a write lock. 
   2. For data deletion of unsealed tsfiles, it is necessary to check whether there is a device containing the data to be deleted in its resource, and whether the time range overlaps. 


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: reviews-unsubscribe@iotdb.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [iotdb] JackieTien97 merged pull request #6743: [IOTDB-3777]Avoid holding sg write lock in the whole processin of deletion

[GitBox <gi...@apache.org>]

JackieTien97 merged PR #6743:
URL: https://github.com/apache/iotdb/pull/6743


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: reviews-unsubscribe@iotdb.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org