blob: 239a033f7e4d4a8c540828d3a29c915154472bf7 [file] [log] [blame]
/*
* Copyright (C) 2010 The Guava Authors
*
* Licensed 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.google.common.collect;
import com.google.caliper.BeforeExperiment;
import com.google.caliper.Benchmark;
import com.google.caliper.Param;
import com.google.common.base.Function;
import java.math.BigInteger;
import java.util.Comparator;
import java.util.PriorityQueue;
import java.util.Queue;
import java.util.Random;
/**
* Benchmarks to compare performance of MinMaxPriorityQueue and PriorityQueue.
*
* @author Sverre Sundsdal
*/
public class MinMaxPriorityQueueBenchmark {
@Param private ComparatorType comparator;
// TODO(kevinb): add 1000000 back when we have the ability to throw
// NotApplicableException in the expensive comparator case.
@Param({"100", "10000"})
private int size;
@Param private HeapType heap;
private Queue<Integer> queue;
private final Random random = new Random();
@BeforeExperiment
void setUp() {
queue = heap.create(comparator.get());
for (int i = 0; i < size; i++) {
queue.add(random.nextInt());
}
}
@Benchmark
void pollAndAdd(int reps) {
for (int i = 0; i < reps; i++) {
// TODO(kevinb): precompute random #s?
queue.add(queue.poll() ^ random.nextInt());
}
}
@Benchmark
void populate(int reps) {
for (int i = 0; i < reps; i++) {
queue.clear();
for (int j = 0; j < size; j++) {
// TODO(kevinb): precompute random #s?
queue.add(random.nextInt());
}
}
}
/**
* Implementation of the InvertedMinMaxPriorityQueue which forwards all calls to a
* MinMaxPriorityQueue, except poll, which is forwarded to pollMax. That way we can benchmark
* pollMax using the same code that benchmarks poll.
*/
static final class InvertedMinMaxPriorityQueue<T> extends ForwardingQueue<T> {
MinMaxPriorityQueue<T> mmHeap;
public InvertedMinMaxPriorityQueue(Comparator<T> comparator) {
mmHeap = MinMaxPriorityQueue.orderedBy(comparator).create();
}
@Override
protected Queue<T> delegate() {
return mmHeap;
}
@Override
public T poll() {
return mmHeap.pollLast();
}
}
public enum HeapType {
MIN_MAX {
@Override
public Queue<Integer> create(Comparator<Integer> comparator) {
return MinMaxPriorityQueue.orderedBy(comparator).create();
}
},
PRIORITY_QUEUE {
@Override
public Queue<Integer> create(Comparator<Integer> comparator) {
return new PriorityQueue<>(11, comparator);
}
},
INVERTED_MIN_MAX {
@Override
public Queue<Integer> create(Comparator<Integer> comparator) {
return new InvertedMinMaxPriorityQueue<>(comparator);
}
};
public abstract Queue<Integer> create(Comparator<Integer> comparator);
}
/**
* Does a CPU intensive operation on Integer and returns a BigInteger Used to implement an
* ordering that spends a lot of cpu.
*/
static class ExpensiveComputation implements Function<Integer, BigInteger> {
@Override
public BigInteger apply(Integer from) {
BigInteger v = BigInteger.valueOf(from);
// Math.sin is very slow for values outside 4*pi
// Need to take absolute value to avoid inverting the value.
for (double i = 0; i < 100; i += 20) {
v =
v.add(
v.multiply(
BigInteger.valueOf(((Double) Math.abs(Math.sin(i) * 10.0)).longValue())));
}
return v;
}
}
public enum ComparatorType {
CHEAP {
@Override
public Comparator<Integer> get() {
return Ordering.natural();
}
},
EXPENSIVE {
@Override
public Comparator<Integer> get() {
return Ordering.natural().onResultOf(new ExpensiveComputation());
}
};
public abstract Comparator<Integer> get();
}
}