zhengchenyu commented on code in PR #1250: URL: https://github.com/apache/incubator-uniffle/pull/1250#discussion_r1366401791
########## client-tez/src/main/java/org/apache/tez/runtime/library/common/shuffle/RemoteFetchedInput.java: ########## @@ -0,0 +1,153 @@ +/* + * 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.tez.runtime.library.common.shuffle; + +import java.io.IOException; +import java.io.InputStream; +import java.io.OutputStream; + +import com.google.common.base.Preconditions; +import org.apache.hadoop.fs.FileSystem; +import org.apache.hadoop.fs.Path; +import org.apache.tez.runtime.library.common.InputAttemptIdentifier; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +public class RemoteFetchedInput extends FetchedInput { + + private static final Logger LOG = LoggerFactory.getLogger(RemoteFetchedInput.class); + + /* + * Output path must be unique, is composed of unique_id, src_id and spill_id. + * unique_id is task attempt id + io context increasing id. + * src_id the source input id, is determined by partition and source task. + * spill_id is the spill id of map task. + * */ + private static final String OUTPUT_FILE_PATTERN = "%s/%s_src_%d_spill_%d.out"; + + private final FileSystem remoteFS; + private final Path tmpOutputPath; + private final Path outputPath; + private final long size; + + public RemoteFetchedInput( + long compressedSize, + InputAttemptIdentifier inputAttemptIdentifier, + FetchedInputCallback callbackHandler, + FileSystem remoteFS, + String remoteSpillBasePath, + String uniqueId, + String appAttemptId) { + super(inputAttemptIdentifier, callbackHandler); + this.size = compressedSize; + this.remoteFS = remoteFS; + String outputFile = + String.format( + OUTPUT_FILE_PATTERN, + appAttemptId, + uniqueId, + this.getInputAttemptIdentifier().getInputIdentifier(), + this.getInputAttemptIdentifier().getSpillEventId()); + this.outputPath = new Path(remoteSpillBasePath, outputFile); + // Files are not clobbered due to the id being appended to the outputPath in the tmpPath, + // otherwise fetches for the same task but from different attempts would clobber each other. + this.tmpOutputPath = outputPath.suffix(String.valueOf(getId())); + } + + @Override + public Type getType() { + return Type.DISK; + } + + @Override + public long getSize() { + return size; + } + + @Override + public OutputStream getOutputStream() throws IOException { + return remoteFS.create(tmpOutputPath); + } + + @Override + public InputStream getInputStream() throws IOException { + return remoteFS.open(outputPath); + } + + public final Path getInputPath() { + if (isState(State.COMMITTED)) { + return this.outputPath; + } + return this.tmpOutputPath; + } + + @Override + public void commit() throws IOException { + if (isState(State.PENDING)) { + setState(State.COMMITTED); + remoteFS.rename(tmpOutputPath, outputPath); + notifyFetchComplete(); + } + } + + @Override + public void abort() throws IOException { + if (isState(State.PENDING)) { + setState(State.ABORTED); + // TODO NEWTEZ Maybe defer this to container cleanup + remoteFS.delete(tmpOutputPath, false); + notifyFetchFailure(); + } + } + + @Override + public void free() { Review Comment: We don’t have to think about thread safety, it's impossible to run in parallel. When write operation is done, then call commit, then send to completed queue, then read or free. Other implement for FetchInput also does not consider thread safety. ########## client-tez/src/main/java/org/apache/tez/runtime/library/common/shuffle/RemoteFetchedInput.java: ########## @@ -0,0 +1,153 @@ +/* + * 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.tez.runtime.library.common.shuffle; + +import java.io.IOException; +import java.io.InputStream; +import java.io.OutputStream; + +import com.google.common.base.Preconditions; +import org.apache.hadoop.fs.FileSystem; +import org.apache.hadoop.fs.Path; +import org.apache.tez.runtime.library.common.InputAttemptIdentifier; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +public class RemoteFetchedInput extends FetchedInput { + + private static final Logger LOG = LoggerFactory.getLogger(RemoteFetchedInput.class); + + /* + * Output path must be unique, is composed of unique_id, src_id and spill_id. + * unique_id is task attempt id + io context increasing id. + * src_id the source input id, is determined by partition and source task. + * spill_id is the spill id of map task. + * */ + private static final String OUTPUT_FILE_PATTERN = "%s/%s_src_%d_spill_%d.out"; + + private final FileSystem remoteFS; + private final Path tmpOutputPath; + private final Path outputPath; + private final long size; + + public RemoteFetchedInput( + long compressedSize, + InputAttemptIdentifier inputAttemptIdentifier, + FetchedInputCallback callbackHandler, + FileSystem remoteFS, + String remoteSpillBasePath, + String uniqueId, + String appAttemptId) { + super(inputAttemptIdentifier, callbackHandler); + this.size = compressedSize; + this.remoteFS = remoteFS; + String outputFile = + String.format( + OUTPUT_FILE_PATTERN, + appAttemptId, + uniqueId, + this.getInputAttemptIdentifier().getInputIdentifier(), + this.getInputAttemptIdentifier().getSpillEventId()); + this.outputPath = new Path(remoteSpillBasePath, outputFile); + // Files are not clobbered due to the id being appended to the outputPath in the tmpPath, + // otherwise fetches for the same task but from different attempts would clobber each other. + this.tmpOutputPath = outputPath.suffix(String.valueOf(getId())); + } + + @Override + public Type getType() { + return Type.DISK; + } + + @Override + public long getSize() { + return size; + } + + @Override + public OutputStream getOutputStream() throws IOException { + return remoteFS.create(tmpOutputPath); + } + + @Override + public InputStream getInputStream() throws IOException { + return remoteFS.open(outputPath); + } + + public final Path getInputPath() { + if (isState(State.COMMITTED)) { + return this.outputPath; + } + return this.tmpOutputPath; + } + + @Override + public void commit() throws IOException { + if (isState(State.PENDING)) { + setState(State.COMMITTED); + remoteFS.rename(tmpOutputPath, outputPath); + notifyFetchComplete(); + } + } + + @Override + public void abort() throws IOException { + if (isState(State.PENDING)) { + setState(State.ABORTED); + // TODO NEWTEZ Maybe defer this to container cleanup + remoteFS.delete(tmpOutputPath, false); + notifyFetchFailure(); + } + } + + @Override + public void free() { Review Comment: We don’t have to think about thread safety, it's impossible to run in parallel. When write operation is done, then call commit, then send to completed queue, then read or free. Other implement for FetchInput also does not consider thread safety. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
