flink-tests/src/test/java/org/apache/flink/test/misc/SuccessAfterNetworkBuffersFailureITCase.java

/*
 * 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.test.misc;

import org.apache.flink.api.java.DataSet;
import org.apache.flink.api.java.ExecutionEnvironment;
import org.apache.flink.api.java.aggregation.Aggregations;
import org.apache.flink.api.java.io.DiscardingOutputFormat;
import org.apache.flink.api.java.operators.DeltaIteration;
import org.apache.flink.api.java.operators.IterativeDataSet;
import org.apache.flink.api.java.tuple.Tuple2;
import org.apache.flink.configuration.Configuration;
import org.apache.flink.configuration.MemorySize;
import org.apache.flink.configuration.TaskManagerOptions;
import org.apache.flink.examples.java.clustering.KMeans;
import org.apache.flink.examples.java.clustering.util.KMeansData;
import org.apache.flink.examples.java.graph.ConnectedComponents;
import org.apache.flink.examples.java.graph.util.ConnectedComponentsData;
import org.apache.flink.runtime.client.JobExecutionException;
import org.apache.flink.runtime.testutils.MiniClusterResourceConfiguration;
import org.apache.flink.test.util.MiniClusterWithClientResource;
import org.apache.flink.util.TestLogger;

import org.junit.ClassRule;
import org.junit.Test;

import static org.apache.flink.util.ExceptionUtils.findThrowableWithMessage;
import static org.junit.Assert.assertTrue;
import static org.junit.Assert.fail;

/**
 * Test that runs an iterative job after a failure in another iterative job. This test validates
 * that task slots in co-location constraints are properly freed in the presence of failures.
 */
public class SuccessAfterNetworkBuffersFailureITCase extends TestLogger {

    private static final int PARALLELISM = 16;

    @ClassRule
    public static final MiniClusterWithClientResource MINI_CLUSTER_RESOURCE =
            new MiniClusterWithClientResource(
                    new MiniClusterResourceConfiguration.Builder()
                            .setConfiguration(getConfiguration())
                            .setNumberTaskManagers(2)
                            .setNumberSlotsPerTaskManager(8)
                            .build());

    private static Configuration getConfiguration() {
        Configuration config = new Configuration();
        config.set(TaskManagerOptions.MANAGED_MEMORY_SIZE, MemorySize.parse("80m"));
        config.set(TaskManagerOptions.NETWORK_MEMORY_MIN, MemorySize.ofMebiBytes(32L));
        config.set(TaskManagerOptions.NETWORK_MEMORY_MAX, MemorySize.ofMebiBytes(32L));
        return config;
    }

    @Test
    public void testSuccessfulProgramAfterFailure() throws Exception {
        ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

        runConnectedComponents(env);

        try {
            runKMeans(env);
            fail("This program execution should have failed.");
        } catch (JobExecutionException e) {
            assertTrue(
                    findThrowableWithMessage(e, "Insufficient number of network buffers")
                            .isPresent());
        }

        runConnectedComponents(env);
    }

    private static void runConnectedComponents(ExecutionEnvironment env) throws Exception {

        env.setParallelism(PARALLELISM);

        // read vertex and edge data
        DataSet<Long> vertices = ConnectedComponentsData.getDefaultVertexDataSet(env).rebalance();

        DataSet<Tuple2<Long, Long>> edges =
                ConnectedComponentsData.getDefaultEdgeDataSet(env)
                        .rebalance()
                        .flatMap(new ConnectedComponents.UndirectEdge());

        // assign the initial components (equal to the vertex id)
        DataSet<Tuple2<Long, Long>> verticesWithInitialId =
                vertices.map(new ConnectedComponents.DuplicateValue<Long>());

        // open a delta iteration
        DeltaIteration<Tuple2<Long, Long>, Tuple2<Long, Long>> iteration =
                verticesWithInitialId.iterateDelta(verticesWithInitialId, 100, 0);

        // apply the step logic: join with the edges, select the minimum neighbor,
        // update if the component of the candidate is smaller
        DataSet<Tuple2<Long, Long>> changes =
                iteration
                        .getWorkset()
                        .join(edges)
                        .where(0)
                        .equalTo(0)
                        .with(new ConnectedComponents.NeighborWithComponentIDJoin())
                        .groupBy(0)
                        .aggregate(Aggregations.MIN, 1)
                        .join(iteration.getSolutionSet())
                        .where(0)
                        .equalTo(0)
                        .with(new ConnectedComponents.ComponentIdFilter());

        // close the delta iteration (delta and new workset are identical)
        DataSet<Tuple2<Long, Long>> result = iteration.closeWith(changes, changes);

        result.output(new DiscardingOutputFormat<Tuple2<Long, Long>>());

        env.execute();
    }

    private static void runKMeans(ExecutionEnvironment env) throws Exception {

        env.setParallelism(PARALLELISM);

        // get input data
        DataSet<KMeans.Point> points = KMeansData.getDefaultPointDataSet(env).rebalance();
        DataSet<KMeans.Centroid> centroids = KMeansData.getDefaultCentroidDataSet(env).rebalance();

        // set number of bulk iterations for KMeans algorithm
        IterativeDataSet<KMeans.Centroid> loop = centroids.iterate(20);

        // add some re-partitions to increase network buffer use
        DataSet<KMeans.Centroid> newCentroids =
                points
                        // compute closest centroid for each point
                        .map(new KMeans.SelectNearestCenter())
                        .withBroadcastSet(loop, "centroids")
                        .rebalance()
                        // count and sum point coordinates for each centroid
                        .map(new KMeans.CountAppender())
                        .groupBy(0)
                        .reduce(new KMeans.CentroidAccumulator())
                        // compute new centroids from point counts and coordinate sums
                        .rebalance()
                        .map(new KMeans.CentroidAverager());

        // feed new centroids back into next iteration
        DataSet<KMeans.Centroid> finalCentroids = loop.closeWith(newCentroids);

        DataSet<Tuple2<Integer, KMeans.Point>> clusteredPoints =
                points
                        // assign points to final clusters
                        .map(new KMeans.SelectNearestCenter())
                        .withBroadcastSet(finalCentroids, "centroids");

        clusteredPoints.output(new DiscardingOutputFormat<Tuple2<Integer, KMeans.Point>>());

        env.execute("KMeans Example");
    }
}