我大约要存储20万条记录。我已经实现了Java客户端,以从hazelcast Map中搜索记录。我没有在预期的时间内得到搜索结果。
一旦我对“ Hazelcast”进行“赞”或“查询”,就需要花费至少400到500毫秒。
可以更改服务器和客户端的配置以提高吞吐量吗?
我已将Java Bean信息与键值一起存储在Map中。我还在一个字段上创建了索引。还实现了身份序列化机制。
服务器端配置(使用XML文件设置服务器):
<map name="app-data">
<in-memory-format>BINARY</in-memory-format>
<statistics-enabled>false</statistics-enabled>
<optimize-queries>false</optimize-queries>
<cache-deserialized-values>INDEX-ONLY</cache-deserialized-values>
<backup-count>0</backup-count>
<async-backup-count>0</async-backup-count>
<time-to-live-seconds>0</time-to-live-seconds>
<max-idle-seconds>0</max-idle-seconds>
<eviction-policy>NONE</eviction-policy>
<max-size policy="PER_NODE">2147483647</max-size>
<eviction-percentage>25</eviction-percentage>
<min-eviction-check-millis>100</min-eviction-check-millis>
<merge-policy batch-size="100">com.hazelcast.map.merge.LatestUpdateMapMergePolicy</merge-policy>
<read-backup-data>false</read-backup-data>
<hot-restart enabled="false">
<fsync>false</fsync>
</hot-restart>
<map-store enabled="false" initial-mode="LAZY">
<write-delay-seconds>0</write-delay-seconds>
<write-batch-size>1</write-batch-size>
</map-store>
<indexes>
<index ordered="true">index</index>
</indexes>
</map>
<data-serializable-factories>
<data-serializable-factory factory-id="1">com.tmobile.services.cacheserver.config.hazelcast.cache.serialization.CacheServerDataSerializableFactory</data-serializable-factory>
客户代码:
ClientConfig config = new ClientConfig();
config.getSerializationConfig()
.addDataSerializableFactory(
CacheServerDataSerializableFactory.FACTORY_ID,
new CacheServerDataSerializableFactory());
config.getNetworkConfig().setAddresses(addresses).setConnectionAttemptPeriod(3000)
.getSocketOptions().setBufferSize(256 * SocketOptions.KILO_BYTE);
HazelcastInstance hazelcastInstance = HazelcastClient.newHazelcastClient(config);
return hazelcastInstance;
我使用的是hazelcast版本3.11。
下面是我用来存储信息的对象。
public class AppData implements IdentifiedDataSerializable {
private static final long serialVersionUID = 1L;
@JsonIgnore
private Long index = 0l;
private Integer appId = 0;
public void setIndex(Integer index) {
this.index = index;
}
public Integer getIndex() {
return this.index;
}
public void setAppId(Integer appId) {
this.appId = appId;
}
public Integer getAppId() {
return this.appId;
}
@Override
public final int hashCode() {
return Objects.hash(this.getAppId());
}
@Override
public void writeData(ObjectDataOutput out) throws IOException {
out.writeLong(index);
out.writeInt(appId);
}
@Override
public void readData(ObjectDataInput in) throws IOException {
this.index = in.readLong();
this.appId = in.readInt();
}
@Override
@JsonIgnore
public int getFactoryId() {
return CacheServerDataSerializableFactory.APP_DATE_FACTORY_TYPE;
}
@Override
@JsonIgnore
public int getId() {
return CacheServerDataSerializableFactory.APP_DATE_ID;
}
}
我在下面的查询/谓词中使用过过滤数据。
Predicate pred = new InPredicate(attribute, (Comparable[]) ((List) filterValue).toArray(new String[0]));
Collection<Object> items = map.values(predicate);
答案 0 :(得分:0)
@Prashant,由于IN
谓词仅命中索引n次并直接获得结果,因此唯一可以改善的地方就是索引复制行为:https://docs.hazelcast.org/docs/3.11.1/manual/html-single/index.html#copying-indexes
您可以在成员config上设置以下属性并查看结果:
<property name="hazelcast.index.copy.behavior">NEVER</property>
请阅读有关索引复制行为的文档。