flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FLINK-8845) Use WriteBatch to improve performance for recovery in RocksDB backend
Date Wed, 07 Mar 2018 18:56:00 GMT

    [ https://issues.apache.org/jira/browse/FLINK-8845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16390026#comment-16390026
] 

ASF GitHub Bot commented on FLINK-8845:
---------------------------------------

Github user bowenli86 commented on a diff in the pull request:

    https://github.com/apache/flink/pull/5650#discussion_r172935214
  
    --- 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");
    +
    +		this.db = rocksDB;
    +		this.options = options;
    +		this.capacity = capacity;
    +		this.batch = new WriteBatch(this.capacity);
    +		this.currentSize = 0;
    +	}
    +
    +	public void put(ColumnFamilyHandle handle, byte[] key, byte[] value) throws RocksDBException
{
    --- End diff --
    
    need synchronization on put() and flush()


> 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)

Mime
View raw message