apex-dev 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] (APEXMALHAR-1963) Add abstract operator for Async Processing
Date Wed, 06 Jan 2016 09:38:39 GMT

    [ https://issues.apache.org/jira/browse/APEXMALHAR-1963?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15085303#comment-15085303
] 

ASF GitHub Bot commented on APEXMALHAR-1963:
--------------------------------------------

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

    https://github.com/apache/incubator-apex-malhar/pull/157#discussion_r48940518
  
    --- Diff: library/src/main/java/com/datatorrent/lib/async/AbstractAsyncProcessor.java
---
    @@ -0,0 +1,357 @@
    +/**
    + * 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 com.datatorrent.lib.async;
    +
    +import java.util.Iterator;
    +import java.util.Queue;
    +import java.util.concurrent.ExecutorService;
    +import java.util.concurrent.Executors;
    +import java.util.concurrent.Future;
    +
    +import org.slf4j.Logger;
    +import org.slf4j.LoggerFactory;
    +
    +import com.google.common.collect.Lists;
    +
    +import com.datatorrent.api.Context;
    +import com.datatorrent.api.DefaultInputPort;
    +import com.datatorrent.api.Operator.IdleTimeHandler;
    +import com.datatorrent.api.annotation.InputPortFieldAnnotation;
    +import com.datatorrent.common.util.BaseOperator;
    +
    +/**
    + * This base operator having a single input and tuple coming from input port are enqueued
for processing asynchronously
    + * by Thread Executors.
    + *
    + * User can configure number of of threads that are required for processing. Defaulting
to single thread which is
    + * different than operator thread.
    + * The user will also configure the timeout for the operations to happen.
    + *
    + * <b>Workflow is as follows:</b>
    + * <ol>
    + *   <li>Tuple of type INPUT is received on input port.</li>
    + *   <li>Tuple gets enqueued into Executors by calling <i>enqueueTupleForProcessing</i>.
    + *       Once can override <i>processTuple</i> method in which case, its
users responsibility to enqueue tuple for
    + *       processing by calling <i>enqueueTupleForProcessing</i>.</li>
    + *   <li>When tuple gets its turn of execution, <i>processTupleAsync</i>
gets called from thread other than
    + *       operator thread.</li>
    + *   <li>Based on maintainTupleOrder parameter, the results are consolidated at
end of every window and respective call
    + *       to <i>handleProcessedTuple</i> will be made. This call will happen
in operator thread.</li>
    + * </ol>
    + *
    + * This operator can be implemented to do async and parallel operations.
    + *
    + * Use case examples:
    + * <ul>
    + *   <li>Parallel reads from external datastore/database system.</li>
    + *   <li>Doing any operations which are long running tasks per tuple but DAG io
should not be blocked.</li>
    + * </ul>
    + *
    + * @param <INPUT> input type of tuple
    + * @param <RESULT> Result of async processing of tuple.
    + * @since 3.3.0
    + */
    +public abstract class AbstractAsyncProcessor<INPUT, RESULT> extends BaseOperator
implements IdleTimeHandler
    +{
    +  private static final Logger logger = LoggerFactory.getLogger(AbstractAsyncProcessor.class);
    +
    +  private transient ExecutorService executor;
    +  private int numProcessors = 1;
    +  private boolean maintainTupleOrder = false;
    +  private long processTimeoutMs = 0;
    +
    +  /**
    +   * This Queue will hold all the tuples that are in process, has completed processing
but yet to be notified for
    +   * processing completion.
    +   */
    +  private Queue<ProcessTuple> waitingTuples = Lists.newLinkedList();
    +
    +  @InputPortFieldAnnotation(optional = true)
    +  public final transient DefaultInputPort<INPUT> input = new DefaultInputPort<INPUT>()
    +  {
    +    @Override public void process(INPUT input)
    +    {
    +      processTuple(input);
    +    }
    +
    +    @Override public void setup(Context.PortContext context)
    +    {
    +      setupInputPort(context);
    +    }
    +  };
    +
    +  /**
    +   * This method is exposed for implementing class so as to take care of input port based
initialization.
    +   * For eg. Retrieve TUPLE_CLASS attribute from input port.
    +   *
    +   * @param context PortContext of input port
    +   */
    +  protected void setupInputPort(Context.PortContext context)
    +  {
    +    // Do nothing. Concrete class may choose to do override this. For eg. reading TUPLE_CLASS
attribute.
    +  }
    +
    +  /**
    +   * Setup method of this operator will initialize the various types of executors possible.
    +   *
    +   * @param context OperatorContext
    +   */
    +  @Override public void setup(Context.OperatorContext context)
    +  {
    +    if (numProcessors == 1) {
    +      this.executor = Executors.newSingleThreadExecutor();
    +    } else if (numProcessors == 0) {
    +      this.executor = Executors.newCachedThreadPool();
    +    } else if (numProcessors > 0) {
    +      this.executor = Executors.newFixedThreadPool(numProcessors);
    +    } else {
    --- End diff --
    
    @Min = 0  for numProcessors? Will save this else case.


> Add abstract operator for Async Processing
> ------------------------------------------
>
>                 Key: APEXMALHAR-1963
>                 URL: https://issues.apache.org/jira/browse/APEXMALHAR-1963
>             Project: Apache Apex Malhar
>          Issue Type: New Feature
>    Affects Versions: 3.3.0
>            Reporter: Chinmay Kolhatkar
>            Assignee: Chinmay Kolhatkar
>
> Create an abstract operator which does following:
> 1) Asynchronously processes the tuples
> 2) Have parallel executions
> 3) Make sure the outbound tuples are ordered similar to inbound tuples if configured.
> 4) Have processing timeout.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message