MySQL 高可用架构.pptx

2016MySQL 高可用架构邵汉成技术中心高可用：冗余与备份高大上的MySQL ClusterMySQL 监控面向写操作的横向伸缩面向读操作的横向伸缩Replication与BinLog需要达成的目标CONTENTS目录实现目标冗余与备份可随时、按需Scale out冗余：当一个或几个Node出现问题时，系统可以正常运行备份：使用备份可以将Node恢复到Crash之前的状态高可用可伸缩Scaling out readsScaling out writes百万订单/天5百亿元/年快速响应监控系统的运行响应快易监控快速响应用户请求，用户体验好Understand the load onour systemKeep a keen eye on any problems大容量驴妈妈系统业务容量目标订单量 100万单每天销售量 500亿元每年高可用：冗余与备份高大上的MySQL ClusterMySQL 监控面向写操作的横向伸缩面向读操作的横向伸缩Replication与BinLog需要达成的目标CONTENTS目录高可用提供冗余与备份冗余冗余：提供多台master和slave，防止单点故障或几个node故障备份：Crash之后可以迅速恢复故障发生时的恢复计划99.999%计划Slave FailuresMaster FailuresDisaster Recovery详细的执行方法方法基本的准备工作：增加新Slaves，移除Slaves，切换MasterMaster Hot StandbyDual MastersSlave PromotionCircular Replication高可用Master Hot Standby 结构高可用Dual Masters 结构Active – Active: 两个Master同时接受写入，并将更新数据同步到另一个master。常见用途：根据用户所在区域，用户访问就近的masterActive – Passive: 写操作写入active master，同时passive master只保持与active master 同步高可用Circular Replication 结构All subscriber data is replicated to all sites, and updates of data are allowed in all data centers.高可用Slave Promotiontraditional method of slave promotionrevised method for promoting a slave高可用：冗余与备份高大上的MySQL ClusterMySQL 监控面向写操作的横向伸缩面向读操作的横向伸缩Replication与BinLog需要达成的目标CONTENTS目录Replication与BinLog机制Replication基本介绍Replication与BinLog机制Binary Log 基本介绍基于Statement的复制基于Row的复制Replication与BinLog机制深入理解 Replication结构Replication与BinLog机制GTID: Global Transaction IDEach transaction on a server is assigned a transaction identifier.assigned to a transaction based on the order in which they committed.The GTID of the transaction is retained regardless of the number of times that the transaction is propagated.高可用：冗余与备份高大上的MySQL ClusterMySQL 监控面向写操作的Scale out面向读操作的Scale outReplication与BinLog需要达成的目标CONTENTS目录面向读操作的Scale out读Scale out 对系统Load的帮助一台服务器容量为10000TPS，支持写操作4000TPS，读操作6000TPS：增加3台Slaves之后：面向读操作的Scale out同步或异步 Replication 的选择Latency (ms)Transaction commit time (ms) Equivalent transactions per secondExample case0.010.14~7,100Samecom puter0.10.5~2,000Small LAN14.1~240Bigger LAN