[ https://issues.apache.org/jira/browse/FLINK-8360?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16374450#comment-16374450 ]
ASF GitHub Bot commented on FLINK-8360: --------------------------------------- Github user StefanRRichter commented on a diff in the pull request: https://github.com/apache/flink/pull/5239#discussion_r170271924 --- Diff: flink-runtime/src/main/java/org/apache/flink/runtime/state/filesystem/FileBasedStateOutputStream.java --- @@ -0,0 +1,155 @@ +/* + * 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.runtime.state.filesystem; + +import org.apache.flink.core.fs.FSDataOutputStream; +import org.apache.flink.core.fs.FileSystem; +import org.apache.flink.core.fs.FileSystem.WriteMode; +import org.apache.flink.core.fs.Path; +import org.apache.flink.runtime.state.CheckpointStreamFactory.CheckpointStateOutputStream; + +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import javax.annotation.Nonnull; + +import java.io.IOException; + +import static org.apache.flink.util.Preconditions.checkNotNull; + +/** + * A {@link CheckpointStateOutputStream} that writes into a specified file and + * returns a {@link FileStateHandle} upon closing. + * + * <p>Unlike the {@link org.apache.flink.runtime.state.filesystem.FsCheckpointStreamFactory.FsCheckpointStateOutputStream}, + * this stream does not have a threshold below which it returns a memory byte stream handle, + * and does not create random files, but writes to a specified file. + */ +public final class FileBasedStateOutputStream extends CheckpointStateOutputStream { + + private static final Logger LOG = LoggerFactory.getLogger(FileBasedStateOutputStream.class); + + // ------------------------------------------------------------------------ + + private final FSDataOutputStream out; + + private final Path path; + + private final FileSystem fileSystem; + + private volatile boolean closed; + + + public FileBasedStateOutputStream(FileSystem fileSystem, Path path) throws IOException { + this.fileSystem = checkNotNull(fileSystem); + this.path = checkNotNull(path); + + this.out = fileSystem.create(path, WriteMode.NO_OVERWRITE); + } + + // ------------------------------------------------------------------------ + // I/O + // ------------------------------------------------------------------------ + + @Override + public final void write(int b) throws IOException { + out.write(b); --- End diff -- The second one does not 4 bytes as the argument would make you think, but extracts the lower byte from the int. From the doc of `OutputStream`: ``` * Writes the specified byte to this output stream. The general * contract for <code>write</code> is that one byte is written * to the output stream. (...) ``` > Implement task-local state recovery > ----------------------------------- > > Key: FLINK-8360 > URL: https://issues.apache.org/jira/browse/FLINK-8360 > Project: Flink > Issue Type: New Feature > Components: State Backends, Checkpointing > Reporter: Stefan Richter > Assignee: Stefan Richter > Priority: Major > Fix For: 1.5.0 > > > This issue tracks the development of recovery from task-local state. The main > idea is to have a secondary, local copy of the checkpointed state, while > there is still a primary copy in DFS that we report to the checkpoint > coordinator. > Recovery can attempt to restore from the secondary local copy, if available, > to save network bandwidth. This requires that the assignment from tasks to > slots is as sticky is possible. > For starters, we will implement this feature for all managed keyed states and > can easily enhance it to all other state types (e.g. operator state) later. -- This message was sent by Atlassian JIRA (v7.6.3#76005)