[
https://issues.apache.org/jira/browse/FLINK-8845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16390632#comment-16390632
]
ASF GitHub Bot commented on FLINK-8845:
---------------------------------------
Github user sihuazhou commented on a diff in the pull request:
https://github.com/apache/flink/pull/5650#discussion_r173049537
--- Diff:
flink-state-backends/flink-statebackend-rocksdb/src/main/java/org/apache/flink/contrib/streaming/state/RocksDBWriteBatchWrapper.java
---
@@ -0,0 +1,86 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.flink.contrib.streaming.state;
+
+import org.apache.flink.util.Preconditions;
+import org.rocksdb.ColumnFamilyHandle;
+import org.rocksdb.RocksDB;
+import org.rocksdb.RocksDBException;
+import org.rocksdb.WriteBatch;
+import org.rocksdb.WriteOptions;
+
+import javax.annotation.Nonnull;
+
+/**
+ * A wrapper class to wrap WriteBatch.
+ */
+public class RocksDBWriteBatchWrapper implements AutoCloseable {
+
+ private final static int MIN_CAPACITY = 100;
+ private final static int MAX_CAPACITY = 10000;
+
+ private final RocksDB db;
+
+ private final WriteBatch batch;
+
+ private final WriteOptions options;
+
+ private final int capacity;
+
+ private int currentSize;
+
+ public RocksDBWriteBatchWrapper(@Nonnull RocksDB rocksDB,
+
@Nonnull WriteOptions options,
+ int
capacity) {
+
+ Preconditions.checkArgument(capacity >= MIN_CAPACITY &&
capacity <= MAX_CAPACITY,
+ "capacity should at least greater than 100");
--- End diff --
About the capacity range, I didn't find a specific value recommend by
RocksDB, but from [FAQ](https://github.com/facebook/rocksdb/wiki/RocksDB-FAQ)
```
Q: What's the fastest way to load data into RocksDB?
...
2. batch hundreds of keys into one write batch
...
```
I found that they use the word `hundreds`.
> Use WriteBatch to improve performance for recovery in RocksDB backend
> ---------------------------------------------------------------------
>
> Key: FLINK-8845
> URL: https://issues.apache.org/jira/browse/FLINK-8845
> Project: Flink
> Issue Type: Improvement
> Components: State Backends, Checkpointing
> Affects Versions: 1.5.0
> Reporter: Sihua Zhou
> Assignee: Sihua Zhou
> Priority: Major
> Fix For: 1.6.0
>
>
> Base on {{WriteBatch}} we could get 30% ~ 50% performance lift when loading
> data into RocksDB.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
