Hi,
I couldn't gather from the javadoc when IntHashMap should be preferred
over a regular HashMap, i.e. how many objects per map, what type of
objects etc.
I wrote a little program (copy pasted at the bottom) to see how much the
size difference was for IntHashMap, java.util.HashMap and
java.util.concurrent.ConcurrentHashMap.
for 1 instance of new org.apache.wicket.Page(){};
for 10 instances of new org.apache.wicket.Page(){};
for 100 instances of new
org.apache.wicket.ajax.form.AjaxFormSubmitTestPage();
Br,
Tuomas
the results are as follows, respectively:
--------------------------------------------------
map implementation: org.apache.wicket.util.collections.IntHashMap --
size of outputstream: 54 bytes.
map implementation: java.util.HashMap -- size of outputstream: 57 bytes.
map implementation: java.util.concurrent.ConcurrentHashMap -- size of
outputstream: 282 bytes.
HashMap is 5.556% larger than IntHashMap.
ConcurrentHashMap is 422.222% larger than IntHashMap.
--
map implementation: org.apache.wicket.util.collections.IntHashMap --
size of outputstream: 369 bytes.
map implementation: java.util.HashMap -- size of outputstream: 399 bytes.
map implementation: java.util.concurrent.ConcurrentHashMap -- size of
outputstream: 624 bytes.
HashMap is 8.130% larger than IntHashMap.
ConcurrentHashMap is 69.106% larger than IntHashMap.
--
map implementation: org.apache.wicket.util.collections.IntHashMap --
size of outputstream: 24272 bytes.
map implementation: java.util.HashMap -- size of outputstream: 24572 bytes.
map implementation: java.util.concurrent.ConcurrentHashMap -- size of
outputstream: 24797 bytes.
HashMap is 1.236% larger than IntHashMap.
ConcurrentHashMap is 2.163% larger than IntHashMap.
----------
/*
* 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.wicket.util.collections;
import java.io.IOException;
import java.util.HashMap;
import java.util.concurrent.ConcurrentHashMap;
import org.apache.wicket.Page;
import org.apache.wicket.util.io.ByteArrayOutputStream;
import org.apache.wicket.util.io.WicketObjectOutputStream;
import org.apache.wicket.util.tester.WicketTester;
public class Tester
{
private static final int NUMBER_OF_PAGES = 1;
public static void main(String[] args) throws Exception
{
int intHashMapSize = writeMap(populatedIntHashMap());
int hashMapSize = writeMap(populatedHashMap());
int concurrentHashMapSize =
writeMap(populatedConcurrentHashMap());
double hashMapIsLargerPercetange = ((Double.valueOf(hashMapSize) /
intHashMapSize) - 1) * 100;
double concurrentHashMapIsLargerPercetange =
((Double.valueOf(concurrentHashMapSize) / intHashMapSize) - 1) * 100;
System.out.printf("HashMap is %.3f%% larger than IntHashMap.%n",
hashMapIsLargerPercetange);
System.out.printf("ConcurrentHashMap is %.3f%% larger than
IntHashMap.%n",
concurrentHashMapIsLargerPercetange);
}
private static int writeMap(Object map) throws IOException
{
ByteArrayOutputStream out = new ByteArrayOutputStream();
WicketObjectOutputStream objectOutputStream = new
WicketObjectOutputStream(out);
objectOutputStream.writeObject(map);
System.out.printf("map implementation: %s -- size of outputstream: %d
bytes.%n",
map.getClass().getName(), out.size());
return out.size();
}
private static HashMap<Integer, Page> populatedHashMap()
{
new WicketTester();
HashMap<Integer, Page> map = new HashMap<Integer, Page>();
for (int i = 0; i < NUMBER_OF_PAGES; i++)
{
map.put(i, createPage());
}
return map;
}
private static ConcurrentHashMap<Integer, Page>
populatedConcurrentHashMap()
{
new WicketTester();
ConcurrentHashMap<Integer, Page> map = new ConcurrentHashMap<Integer,
Page>();
for (int i = 0; i < NUMBER_OF_PAGES; i++)
{
map.put(i, createPage());
}
return map;
}
private static IntHashMap<Page> populatedIntHashMap()
{
new WicketTester();
IntHashMap<Page> map = new IntHashMap<Page>();
for (int i = 0; i < NUMBER_OF_PAGES; i++)
{
map.put(i, createPage());
}
return map;
}
private static Page createPage()
{
Page page = new Page()
{
};
page.detach();
return page;
}
}
Igor Vaynberg wrote:
exactly what the javadoc says, its a map that does not need to store a
key object. so it is smaller than a regular hashmap when serialized or
kept in memory.
-igor
On Thu, Feb 12, 2009 at 1:40 PM, Tuomas Kärkkäinen
<[email protected]> wrote:
Hi,
I was looking at IntHashMap in wicket trunk.
I couldn't figure out what it does.
The implementation of hashcode for Integer is value, and the autoboxing of
int to Integer comes out to Integer.valueOf(int) which is implemented as an
array lookup for Integers in the range of -128 to 127, and beyond that it's
just new Integer(int). Page ids start at zero for each page in each session
so most of the time they will be between zero and 127.
Br,
Tuomas
