# ElasticSearch 最佳实践

DevFeed: [ElasticSearch 最佳实践](<https://devfeed.tech/articles/elasticsearch-40980.md>)

Original publisher: [Read original article](<https://blog.joway.io/posts/elasticsearch-bp/>)

Author: Joway

Published: 2017-05-28T00:00:00Z

Content type: tutorial

Language: zh

Sources: [Random Thoughts](<https://devfeed.tech/sources/random-thoughts.md>)

Topics: [elasticsearch](<https://devfeed.tech/topics/elasticsearch.md>), [Amazon Web Services](<https://devfeed.tech/topics/aws.md>), [kibana](<https://devfeed.tech/topics/kibana.md>), [Ansible](<https://devfeed.tech/topics/ansible.md>), [Kafka](<https://devfeed.tech/topics/kafka.md>)

Tags: [ansible](<https://devfeed.tech/tags/ansible.md>), [aws](<https://devfeed.tech/tags/aws.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [elasticsearch](<https://devfeed.tech/tags/elasticsearch.md>), [kafka](<https://devfeed.tech/tags/kafka.md>), [kibana](<https://devfeed.tech/tags/kibana.md>), [ssd](<https://devfeed.tech/tags/ssd.md>), [tech](<https://devfeed.tech/tags/tech.md>)

## AI overview

A practical guide to tuning and operating Elasticsearch clusters, covering hardware and JVM memory settings, CPU behavior during queries and force merges, swap, node and cluster configuration, shard recovery, rolling restarts, and logging pipelines. It also discusses using Kafka as a buffer and Ansible for consistent cluster management.

## Source excerpt

Elasticsearch 是一个需要不停调参数的庞然大物 , 从其自身的设置到JVM层面, 有着无数的参数需要根据业务的变化进行调整。最近采用3台 AWS r3.2xlarge , 32GB, 4核, 构建了一套日均日志量过亿的 EFK 套件。经过不停地查阅文档进行调整优化 , 目前日常CPU占用只在30% , 大部分 Kibana 内的查询都能在 5s ~ 15s 内完成。