在開發中大型Java軟件項目時，很多Java架構師都會遇到數據庫讀寫瓶頸，如果你在系統架構時並沒有將緩存策略考慮進去，或者並沒有選擇更優的 緩存策略，那麼到時候重構起來將會是一個噩夢。本文主要是分享了5個常用的Java分布式緩存框架，這些緩存框架支持多台服務器的緩存讀寫功能，可以讓你 的緩存系統更容易擴展。
Ehcache是一個Java實現的開源分布式緩存框架，EhCache 可以有效地減輕數據庫的負載，可以讓數據保存在不同服務器的內存中，在需要數據的時候可以快速存取。同時EhCache 擴展非常簡單，官方提供的Cache配置方式有好幾種。你可以通過聲明配置、在xml中配置、在程序裡配置或者調用構造方法時傳入不同的參數。
<ehcache>
<diskStore path=”java.io.tmpdir”/>
<defaultCache
maxElementsInMemory=”10000″
eternal=”false”
timeToIdleSeconds=”120″
timeToLiveSeconds=”120″
overflowToDisk=”true”
maxElementsOnDisk=”10000000″
diskPersistent=”false”
diskExpiryThreadIntervalSeconds=”120″
memoryStoreEvictionPolicy=”LRU”
/>
</ehcache>
總結
在同類的Java緩存框架中，Ehcache配置相對簡單，也比較容易上手，最大的優勢是它支持分布式緩存
Cacheonix同樣也是一個基於Java的分布式集群緩存系統，它同樣可以幫助你實現分布式緩存的部署。
官方網站：http://www.cacheonix.com/
Cacheonix的特點
Cacheonix的架構圖
<?xml version ="1.0"?>
<cacheonix xmlns="http://www.cacheonix.com/schema/configuration"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://www.cacheonix.com/schema/configuration http://www.cacheonix.com/schema/cacheonix-config-2.0.xsd">
<server>
<listener>
<tcp port="8879" buffer="128k"/>
</listener>
<broadcast>
<multicast multicastAddress="225.0.1.2" multicastPort="9998" multicastTTL="0"/>
</broadcast>
<partitionedCache name="customer.cache">
<store>
<lru maxElements="10000" maxBytes="10mb"/>
<expiration idleTime="120s"/>
</store>
</partitionedCache>
<partitionedCache name="invoice.cache">
<store>
<lru maxElements="10000" maxBytes="10mb"/>
<expiration idleTime="120s"/>
</store>
</partitionedCache>
<partitionedCache name="search.results.cache">
<store>
<lru maxBytes="5mb"/>
</store>
</partitionedCache>
</server>
</cacheonix>
Cacheonix緩存的存取
從配置中獲取Cacheonix實例
/**
* Tester for CacheManager.
*/
public final class CacheonixTest extends TestCase {
private Cacheonix cacheonix;
/**
* Tests getting an instance of CacheManager using a default Cacheonix configuration.
*/
public void testGetInstance() {
assertNotNull("Cacheonix created in setUp() method should not be null", cacheonix);
}
/**
* Sets up the fixture. This method is called before a test is executed.
* <p/>
* Cacheonix receives the default configuration from a <code>cacheonix-config.xml</code> found in a class path or
* using a file that name is defined by system parameter <code>cacheonix.config.xml<code>.
*/
protected void setUp() throws Exception {
super.setUp();
// Get Cacheonix using a default Cacheonix configuration. The configuration
// is stored in the conf/cacheonix-config.xml
cacheonix = Cacheonix.getInstance();
}
/**
* Tears down the fixture. This method is called after a test is executed.
*/
protected void tearDown() throws Exception {
// Cache manager has be be shutdown upon application exit.
// Note that call to shutdown() here uses unregisterSingleton
// set to true. This is necessary to support clean restart on setUp()
cacheonix.shutdown(ShutdownMode.GRACEFUL_SHUTDOWN, true);
cacheonix = null;
super.tearDown();
}
}
讀取緩存
Cacheonix cacheonix = Cacheonix.getInstance();
Cache<String, String> cache = cacheonix.getCache("my.cache");
String cachedValue = cache.get("my.key");
設置緩存
Cacheonix cacheonix = Cacheonix.getInstance();
Cache<String, String> cache = cacheonix.getCache("my.cache");
String replacedValue = cache.put("my.key", "my.value");
刪除緩存
Cacheonix cacheonix = Cacheonix.getInstance();
Cache<String, String> cache = cacheonix.getCache("my.cache");
String removedValue = cache.remove("my.key");
總結
Cacheonix作為一款開源的分布式緩存框架，可以滿足中型企業規模的系統架構，對提升系統性能有非常棒的作用。
ASimpleCache是一款基於Android的輕量級緩存框架，它只有一個Java文件，ASimpleCache基本可以緩存常用的Android對象，包括普通字符串、JSON對象、經過序列化的Java對象、字節數組等。
ACache mCache = ACache.get(this);
mCache.put("test_key1", "test value");
mCache.put("test_key2", "test value", 10);//保存10秒，如果超過10秒去獲取這個key，將為null
mCache.put("test_key3", "test value", 2 * ACache.TIME_DAY);//保存兩天，如果超過兩天去獲取這個key，將為null
獲取緩存數據：
ACache mCache = ACache.get(this);
String value = mCache.getAsString("test_key1");
總結
ASimpleCache的作者是國人，代碼托管在Github上，也用過ASimpleCache的同學可以分享一下使用心得，為開源事業貢獻一份力量。
JBoss Cache是一款基於Java的事務處理緩存系統，它的目標是構建一個以Java框架為基礎的集群解決方案，可以是服務器應用，也可以是Java SE應用。
TreeCache tree = new TreeCache();
然後是讀進配置文件
PropertyConfigurator config = new PropertyConfigurator();
config.configure("配置文件.xml");
然後開始服務
Tree.startService();
因為Tree的結構是用NODE來Access的，TreeCache這裡就很簡單的用：
/level1/level2/node1 來表示兩級Tree下面的Node1。
現在我們添加幾個要Cache的對象。
Tree.put("/level1/level2/node1", "key1", "value1");
String[] array = { "1", "2", "3", "4" }
Tree.put("/level3/array/", "myarray", array);
大家可以看到，TreeCache裡面可以存儲任何種類的對象，包括所有復雜對象。
讀取對象就很方便了，
String s = (String)Tree.get("/level1/level2/node1/", "key1");
value1就讀出來了。
同理：
String[] sarr = (String[]) Tree.get("/level3/array/","myarray");
System.out.println(sarr[1]) 會顯示2
最後停止服務：
Tree.stopService();
JBoss Cache的FileCacheLoader示例
首先創建一個FileCache類封裝JBoss Cache的相關操作，如下：
package com.javaeye.terrencexu.jbosscache;
import java.io.File;
import java.util.Map;
import org.jboss.cache.Cache;
import org.jboss.cache.DefaultCacheFactory;
import org.jboss.cache.Fqn;
import org.jboss.cache.Node;
import org.jboss.cache.config.CacheLoaderConfig;
import org.jboss.cache.config.Configuration;
import org.jboss.cache.loader.FileCacheLoader;
import org.jboss.cache.loader.FileCacheLoaderConfig;
/**
* <p>
* This is demo to illustrate how to use the JBoss Cache to cache your
* frequently accessed Java objects in order to dramatically improve
* the performance of your applications. This makes it easy to remove
* data access bottlenecks, such as connecting to a database.
* </p>
* <p>
* As a rule of thumb, it is recommended that the FileCacheLoader not
* be used in a highly concurrent, transactional or stressful environment,
* ant its use is restricted to testing.
* </p>
*
* @author TerrenceX
*
* @param <T>
*/
public class FileCache<T> {
/**
* The JBoss Cache, used to cache frequently accessed Java objects.
*/
private Cache<String, T> cache;
/**
* @constructor
* @param fsCacheLoaderLocation The file system location to store the cache
*/
public FileCache(File fsCacheLoaderLocation) {
cache = initCache(fsCacheLoaderLocation);
}
/**
* Create a Cache and whose cache loader type is File Cache Loader
*
* @param fsCacheLoaderLocation The file position used to store the cache.
*
* @return Cache
*/
public Cache<String, T> initCache(File fsCacheLoaderLocation) {
// initiate a FileCacheLoader instance
FileCacheLoader fsCacheLoader = new FileCacheLoader();
// prepare the file cache loader configuration file for File Cache Loader
FileCacheLoaderConfig fsCacheLoaderConfig = new FileCacheLoaderConfig();
fsCacheLoaderConfig.setLocation(fsCacheLoaderLocation.toString());
fsCacheLoaderConfig.setCacheLoader(fsCacheLoader);
// set configuration to File Cache Loader
fsCacheLoader.setConfig(fsCacheLoaderConfig);
// prepare the configuration for Cache
Configuration config = new Configuration();
config.setCacheLoaderConfig(new CacheLoaderConfig());
config.getCacheLoaderConfig().addIndividualCacheLoaderConfig(fsCacheLoaderConfig);
// create a Cache through the default cache factory
return new DefaultCacheFactory<String, T>().createCache(config);
}
/**
* Add a new node into the tree-node hierarchy
*
* @param fqn Full Qualified Name for the new node
* @return
*/
public Node<String, T> addNode(Fqn<String> fqn) {
return cache.getRoot().addChild(fqn);
}
/**
* Remove a specified node from the tree-node hierarchy
*
* @param fqn Full Qualified Name for the specified node
*/
public void removeNode(Fqn<String> fqn) {
cache.removeNode(fqn);
}
/**
* Add node information to the specified node.
*
* @param fqn Full Qualified Name for the specified node
* @param key The key of the node information
* @param value The value of the node information
*/
public void addNodeInfo(Fqn<String> fqn, String key, T value) {
cache.put(fqn, key, value);
}
/**
* Batch add node information to the specified node.
*
* @param fqn Full Qualified Name for the specified node
* @param infos Node informations map
*/
public void addNodeInfos(Fqn<String> fqn, Map<String, T> infos) {
cache.put(fqn, infos);
}
/**
* Get node information from the specified node.
*
* @param fqn Full Qualified Name for the specified node
* @param key The key of the node information
* @return
*/
public T getNodeInfo(Fqn<String> fqn, String key) {
return cache.get(fqn, key);
}
/**
* Remove node information from the specified node.
*
* @param fqn Full Qualified Name for the specified node
* @param key The key of the node information
*/
public void removeNodeInfo(Fqn<String> fqn, String key) {
cache.remove(fqn, key);
}
}
下面是一個測試案例：
package com.javaeye.terrencexu.jbosscache;
import java.io.File;
import org.jboss.cache.Fqn;
public class Main {
public static void main(String[] args) {
FileCache<String> fileCache = new FileCache<String>(new File("d:\\tmp"));
Fqn<String> jimmyFqn = Fqn.fromString("/com/manager/jimmy");
Fqn<String> hansonFqn = Fqn.fromString("/com/developer/hanson");
fileCache.addNode(jimmyFqn);
fileCache.addNode(hansonFqn);
fileCache.addNodeInfo(jimmyFqn, "en-name", "Jimmy Zhang");
fileCache.addNodeInfo(jimmyFqn, "zh-name", "Zhang Ji");
fileCache.addNodeInfo(hansonFqn, "en-name", "Hanson Yang");
fileCache.addNodeInfo(hansonFqn, "zh-name", "Yang Kuo");
String enName = fileCache.getNodeInfo(hansonFqn, "en-name");
System.out.println(enName);
}
}
運行結果如下：
- JBossCache MBeans were successfully registered to the platform mbean server. - JBoss Cache version: JBossCache 'Malagueta' 3.2.5.GA Hanson Yang
生成的緩存文件目錄結構如下：
D:/tmp/com.fdb/manage.fdb/jimmy.fdb/data.dat D:/tmp/com.fdb/developer.fdb/hanson.fdb/data.dat
總結
JBoss Cache還有更多的用法，如果你的系統遇到數據庫瓶頸問題，可以考慮使用JBoss Cache來解決。
Voldemort是一款基於Java開發的分布式鍵-值緩存系統，像JBoss Cache一樣，Voldemort同樣支持多台服務器之間的緩存同步，以增強系統的可靠性和讀取性能。
Voldemort邏輯架構圖
Voldemort物理架構圖
Voldemort的配置方式
集群配置文件：
<cluster>
<!-- The name is just to help users identify this cluster from the gui -->
<name>mycluster</name>
<zone>
<zone-id>0</zone-id>
<proximity-list>1</proximity-list>
<zone>
<zone>
<zone-id>1</zone-id>
<proximity-list>0</proximity-list>
<zone>
<server>
<!-- The node id is a unique, sequential id beginning with 0 that identifies each server in the cluster-->
<id>0</id>
<host>vldmt1.prod.linkedin.com</host>
<http-port>8081</http-port>
<socket-port>6666</socket-port>
<admin-port>6667</admin-port>
<!-- A list of data partitions assigned to this server -->
<partitions>0,1,2,3</partitions>
<zone-id>0</zone-id>
</server>
<server>
<id>1</id>
<host>vldmt2.prod.linkedin.com</host>
<http-port>8081</http-port>
<socket-port>6666</socket-port>
<admin-port>6667</admin-port>
<partitions>4,5,6,7</partitions>
<zone-id>1</zone-id>
</server>
</cluster>
數據存儲方式配置文件：
<stores>
<store>
<name>test</name>
<replication-factor>2</replication-factor>
<preferred-reads>2</preferred-reads>
<required-reads>1</required-reads>
<preferred-writes>2</preferred-writes>
<required-writes>1</required-writes>
<persistence>bdb</persistence>
<routing>client</routing>
<routing-strategy>consistent-routing</routing-strategy>
<key-serializer>
<type>string</type>
<schema-info>utf8</schema-info>
</key-serializer>
<value-serializer>
<type>json</type>
<schema-info version="1">[{"id":"int32", "name":"string"}]</schema-info>
<compression>
<type>gzip<type>
</compression>
</value-serializer>
</store>
</stores>
Voldemort的使用示例
value = store.get(key) store.put(key, value) store.delete(key)
總結
Voldemort是分布式緩存系統，因此可以應用在中大型的軟件項目中，性能方面也都還不錯。
