azagrebin commented on a change in pull request #6777: [FLINK-10461] [State
Backends, Checkpointing] Speed up download files when restore from DFS
URL: https://github.com/apache/flink/pull/6777#discussion_r236186387
##########
File path:
flink-core/src/main/java/org/apache/flink/configuration/CheckpointingOptions.java
##########
@@ -115,4 +115,12 @@
.defaultValue(1024)
.withDescription("The minimum size of state data files.
All state chunks smaller than that are stored" +
" inline in the root checkpoint metadata
file.");
+
+ /**
+ * The thread numbers used to download files from DFS in
RocksDBStateBackend.
+ */
+ public static final ConfigOption<Integer> CHECKPOINT_RESTORE_THREAD_NUM
= ConfigOptions
+ .key("state.checkpoint.restore.thread.num")
+ .defaultValue(-1)
Review comment:
I think default value 1 was reasonable before, that is what we have now.
-1 looks rather as undefined value.
I would also consider to change option prefix to contain `rocksdb` in it:
`state.backend.rocksdb.checkpoint.restore.thread.num`
and to move the option to
`org.apache.flink.contrib.streaming.state.RocksDBOptions`.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
