/*
* 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.giraph.io.formats;
import org.apache.giraph.conf.ImmutableClassesGiraphConfiguration;
import org.apache.giraph.partition.PartitionUtils;
import org.apache.giraph.partition.SimpleLongRangePartitionerFactory;
import org.apache.giraph.worker.WorkerInfo;
import java.util.Collections;
import java.util.List;
import java.util.Random;
/**
* Helper class to generate pseudo-random local edges.
*/
public class PseudoRandomLocalEdgesHelper {
/** Minimum ratio of partition-local edges. */
private float minLocalEdgesRatio;
/** Whether we're using range-partitioning or hash-partitioning */
private boolean usingRangePartitioner;
/** Total number of vertices. */
private long numVertices;
/** Total number of partitions. */
private int numPartitions;
/** Average partition size. */
private long partitionSize;
/**
* Constructor.
*
* @param numVertices Total number of vertices.
* @param minLocalEdgesRatio Minimum ratio of local edges.
* @param conf Configuration.
*/
public PseudoRandomLocalEdgesHelper(long numVertices,
float minLocalEdgesRatio,
ImmutableClassesGiraphConfiguration conf)
{
this.minLocalEdgesRatio = minLocalEdgesRatio;
this.numVertices = numVertices;
usingRangePartitioner =
SimpleLongRangePartitionerFactory.class.isAssignableFrom(
conf.getGraphPartitionerClass());
int numWorkers = conf.getMaxWorkers();
List<WorkerInfo> workerInfos = Collections.nCopies(numWorkers,
new WorkerInfo());
numPartitions = PartitionUtils.computePartitionCount(workerInfos,
numWorkers, conf);
partitionSize = numVertices / numPartitions;
}
/**
* Generate a destination vertex id for the given source vertex,
* using the desired configuration for edge locality and the provided
* pseudo-random generator.
*
* @param sourceVertexId Source vertex id.
* @param rand Pseudo-random generator.
* @return Destination vertex id.
*/
public long generateDestVertex(long sourceVertexId, Random rand) {
long destVertexId;
if (rand.nextFloat() < minLocalEdgesRatio) {
if (usingRangePartitioner) {
int partitionId = Math.min(numPartitions - 1,
(int) (sourceVertexId / partitionSize));
destVertexId = partitionId * partitionSize +
(Math.abs(rand.nextLong()) % partitionSize);
} else {
int partitionId = (int) sourceVertexId % numPartitions;
destVertexId = partitionId +
numPartitions * (Math.abs(rand.nextLong()) % partitionSize);
}
} else {
destVertexId = Math.abs(rand.nextLong()) % numVertices;
}
return destVertexId;
}
}