You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@flink.apache.org by rm...@apache.org on 2021/01/18 11:31:30 UTC
[flink-web] 07/10: the JVM heap
This is an automated email from the ASF dual-hosted git repository.
rmetzger pushed a commit to branch asf-site
in repository https://gitbox.apache.org/repos/asf/flink-web.git
commit 031bf5acdcd4a08b93716f1174faa878004cee58
Author: qinjunjerry <11...@users.noreply.github.com>
AuthorDate: Tue Jan 12 14:26:42 2021 +0100
the JVM heap
Co-authored-by: morsapaes <ma...@gmail.com>
---
_posts/2020-12-20-rocksdb.md | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/_posts/2020-12-20-rocksdb.md b/_posts/2020-12-20-rocksdb.md
index 17e26d6..071a9c0 100644
--- a/_posts/2020-12-20-rocksdb.md
+++ b/_posts/2020-12-20-rocksdb.md
@@ -47,7 +47,7 @@ The command is for Linux only. For other operating systems, please refer to thei
In addition to RocksDBStateBackend, Flink has two other built-in state backends: MemoryStateBackend and FsStateBackend. They both are heap-based, as in-flight state is stored in the JVM heap. For the moment being, let’s ignore MemoryStateBackend, as it is intended only for **local developments** and **debugging**, not for production use.
-With RocksDBStateBackend, in-flight state is first written into off-heap/native memory, and then flushed to local disks when a configured threshold is reached. This means that RocksDBStateBackend can support state larger than the total configured heap capacity. The amount of state that you can store in RocksDBStateBackend is only limited by the amount of **disk space** available across the entire cluster. In addition, since RocksDBStateBackend doesn’t use JVM heap to store in-flight sta [...]
+With RocksDBStateBackend, in-flight state is first written into off-heap/native memory, and then flushed to local disks when a configured threshold is reached. This means that RocksDBStateBackend can support state larger than the total configured heap capacity. The amount of state that you can store in RocksDBStateBackend is only limited by the amount of **disk space** available across the entire cluster. In addition, since RocksDBStateBackend doesn’t use the JVM heap to store in-flight [...]
On top of full, self-contained state snapshots, RocksDBStateBackend also supports [incremental checkpointing](https://ci.apache.org/projects/flink/flink-docs-stable/ops/state/large_state_tuning.html#incremental-checkpoints) as a performance tuning option. An incremental checkpoint stores only the changes that occurred since the latest completed checkpoint. This dramatically reduces checkpointing time in comparison to performing a full snapshot. RocksDBStateBackend is currently the only s [...]