与spring jpa相比，内存数据网格中的Hazelcast速度非常慢

Question

对于我们的java项目，我们使用hazelcast服务器和客户端模型在加载应用程序时缓存数据库中的数据。这是我的hazelcast.xml

<hazelcast>
<group>
    <name>dev</name>
    <password>dev-pass</password>
</group>
<management-center enabled="true">http://localhost:8080/mancenter</management-center>
<network>
    <port auto-increment="true">5701</port>
    <join>
        <multicast enabled="false"></multicast>
        <tcp-ip enabled="true">
            <member>172.22.3.74</member>
            <!--<interface>127.0.0.1</interface>-->
        </tcp-ip>        </join>
    <interfaces enabled="false">
        <interface>10.10.1.*</interface>
    </interfaces>
    <symmetric-encryption enabled="false">           
        <algorithm>PBEWithMD5AndDES</algorithm>
        <!-- salt value to use when generating the secret key -->
        <salt>thesalt</salt>
        <!-- pass phrase to use when generating the secret key -->
        <password>thepass</password>
        <!-- iteration count to use when generating the secret key -->
        <iteration-count>19</iteration-count>
    </symmetric-encryption>
</network>
<properties>
    <property name="hazelcast.http.healthcheck.enabled">true</property>
    <property name="hazelcast.health.monitoring.delay.seconds">5</property>
    <property name="hazelcast.health.monitoring.level">NOISY</property>
</properties>


<map name="scripMasterMap">
    <near-cache name="default">
        <in-memory-format>BINARY</in-memory-format>
        <time-to-live-seconds>300</time-to-live-seconds>
        <max-idle-seconds>100</max-idle-seconds>
    </near-cache>       
    <backup-count>1</backup-count>       
    <time-to-live-seconds>8000</time-to-live-seconds>       
    <max-idle-seconds>3000</max-idle-seconds>       
    <eviction-policy>LRU</eviction-policy>       
    <eviction-percentage>25</eviction-percentage>       
    <merge-policy>hz.ADD_NEW_ENTRY</merge-policy>
</map>

​

下面是hazelcast服务器代码。

RefData.java：

HazelcastInstance in = Hazelcast.newHazelcastInstance();
IMap<Integer, ScripMaster> map = in.getMap("scripMasterMap");
for(ScripMaster scripMaster: scripMasterService.getAllScripMasterList()) {
scripMasterMap.put(scripMaster.getAllToken(), scripMaster);
}

以下是hazelcast客户端代码：

ClientConfig cl = new ClientConfig();
HazelcastInstance client = Hazelcast.newHazelcastClient(cl);
IMap<Object, Object> map = cl.getMap("scripMasterMap");
return map;

下面是我的方法，它从hazelcast客户端地图返回数据。

public Set<String> getAssetByInstrumentName(
   IMap<Integer, ScripMaster> map, String instrument) {

  Set<String> myset = new HashSet<String>();
  Predicate namePredicate = Predicates.equal("instrument", instrument);
  Collection<ScripMaster> assets = map.values(namePredicate );
  for(Iterator<ScripMaster> it = assets.iterator(); it.hasNext();) {
    ScripMaster scripmaster = it.next();
    myset.put(scripmaster.getAsset());
  }
  return myset;
}

这是简单的代码。为了让客户端可以在内存中获得数据，是否缺少任何需要做的事情？这个部分现在已经解决了。

public Set<String> getAllCommodity(IMap<Integer, ScripMaster> map) {
Set<String> commoditySet = new HashSet<String>();
Aggregator<Map.Entry<Integer, ScripMaster>, Set<String>> aggregation = new CommodityAggregator();
PropertyExtractor<ScripMaster, String> propertyExtractor = new CommodityPropertyExtractor();
commoditySet = map.aggregate(aggregation);   
//        Aggregators.distinct("scripMaster.commodity");
//        Aggregators.distinct();
LOG.info("Number of commodities loaded::" + commoditySet.size());
return commoditySet;
}

public class CommodityAggregator extends Aggregator <Map.Entry<Integer, ScripMaster>, Set<String>> {

Set<String> commoditySet = new HashSet<String>();

@Override
public void accumulate(Map.Entry<Integer, ScripMaster> input) {

    commoditySet.add(input.getValue().getCommodity());
}

@Override
public void combine(Aggregator aggregator) {
}
@Override
public Set<String> aggregate() {
    return commoditySet;
}

}

Answer 1

在instrument字段上添加索引。