我正在使用8个节点的elasticsearch集群。数据大小约为1.5TB,如果进行并行请求,则群集响应速度非常慢。
该机的容量为500GB SSD硬盘和64GB内存。每台计算机上的堆大小也增加到30GB。 在日志文件中。我找到了这些日志:
100 org.elasticsearch.transport.RemoteTransportException:[Hostname] [0.0.0.0:9300] [indices:data / read / search [phase / query]] 101引起:org.elasticsearch.common.util.concurrent.EsRejectedExecutionException:在EsThreadPoolExecutor上拒绝执行org.ela sticsearch.transport.TcpTransport$RequestHandler@8ee60b [search,queue capacity = 1000,org.elasticsearch.common.util.concurrent .EsThreadPoolExecutor @ c8c87d4 [正在运行,池大小= 13,活动线程= 13,排队任务= 1000,已完成任务= 46331944]] 102 org.elasticsearch.common.util.concurrent.EsAbortPolicy.rejectedExecution(EsAbortPolicy.java:50)〜[elasti csearch-5.1.1.jar:5.1.1] 103.在java.util.concurrent.ThreadPoolExecutor.reject(ThreadPoolExecutor.java:830)〜[?:1.8.0_144] 104.在java.util.concurrent.ThreadPoolExecutor.execute(ThreadPoolExecutor.java:1379)〜[?:1.8.0_144] 105 org.elasticsearch.common.util.concurrent.EsThreadPoolExecutor.doExecute(EsThreadPoolExecutor.java:94)〜[elasticsearch-5.1.1.jar:5.1.1] 106 org.elasticsearch.common.util.concurrent.EsThreadPoolExecutor.execute(EsThreadPoolExecutor.java:89)〜[el asticsearch-5.1.1.jar:5.1.1] 107 at org.elasticsearch.transport.TcpTransport.handleRequest(TcpTransport.java:1343)[elasticsearch-5.1.1.jar:5.1.1] 108 org.elasticsearch.transport.TcpTransport.messageReceived(TcpTransport.java:1242)[elasticsearch-5.1.1.ja r:5.1.1] 109 org.elasticsearch.transport.netty4.Netty4MessageChannelHandler.channelRead(Netty4MessageChannelHandler.j ava:74)[transport-netty4-5.1.1.jar:5.1.1] 110 at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:373)[netty-transport-4.1.6.Final.jar:4.1.6.Final] 111 at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:359)[netty-transport-4.1.6.Final.jar:4.1.6.Final] 112 at io.netty.channel.AbstractChannelHandlerContext.fireChannelRead(AbstractChannelHandlerContext.java:351)[netty-transport-4.1.6.Final.jar:4.1.6.Final] 113 at io.netty.handler.codec.ByteToMessageDecoder.fireChannelRead(ByteToMessageDecoder.java:293)[netty-codec- 4.1.6.Final.jar:4.1.6.Final]
有没有办法优化响应时间。