Management Scaling with Large Number of Connections Routing Traffic to Right “Shard” Read-Write Splitting Load Management Avoiding “Dead” Nodes © 2018 Percona. 33 Options in MySQL © 2018 Percona Solution for MySQL ProxySQL © 2018 Percona. 55 • 100% OpenSource • Well Supported • Read-Write Splitting • Caching • Query Rewrites • Load Balancing • 10k+ Connection Support ProxySQL © 2018 Percona Tungsten Replicator • MySQL to Tarantool Replication • MySQL to ClickHouse Replication Custom Binary Log Consumers © 2018 Percona. 65 Summary Lots of Options Exist for Building Distributed Architectures

query © 2018 Percona. 7 Clickhouse Answers • 10x+ times space reduction compared to Raw Text Log Files High Compression (column store + LZ4) • Typically 100x faster than MySQL on Single com/Altinity/clicktail Created by my friends at Altinity Replaced HoneyComb “Sender” with Clickhouse Added Audit Log Support Meet ClickTail © 2018 Percona. 14 Installing ClickTail • curl -s https://packagecloud To run as a service © 2018 Percona. 15 MySQL Logs Primer General Query Log Binary Log Slow Query Log Audit Log © 2018 Percona. 16 MySQL Audit Logs to ClickHouse © 2018 Percona. 17 When

finalizes the installation for products that do not require configuration. It enables you to copy the log to a clipboard and to start certain applications, such as MySQL Workbench and MySQL Shell. Click Finish defining custom file paths for the error log, general log, slow query log (including the configuration of seconds it requires to execute a query), and the binary log. During the configuration process, click in this group should be limited and managed. Windows requires a newly added member to first log out and then log in again to join a local group. • Full access to all users (NOT RECOMMENDED). This option

Separate log_dir and datadir. All storage types benefit from this. For both Percona and MySQL Server, it means setting up the parameters from Appendix A marked with either or <log storage> tmpdir /tmp /<log storage>/mysql_log lc‐messages‐dir /usr/share/mysql explicit_defaults_for_timestamp innodb_log_group_home_dir /<log storage>/mysql_log Best Practices for innodb_undo_directory /<log storage>/mysql_log innodb_buffer_pool_size 3GB 12GB innodb_thread_concurrency 0 innodb_temp_data_file_path '../../../<log storage>/mysql_log/ibtmp1:72 M:autoextend'

1024M innodb_flush_log_at_trx _commit 1 0 0 代表日志只大约每秒写入日志文件并且日志文件 刷新到磁盘 ; 1 为执行完没执行一条 SQL 马上 commit; 2 代表日志写入日志文件在每次提交 后 , 但是日志文件只有大约每秒才会刷新到磁盘上 . 对速度影响比较大，同时也关系数据完整性 innodb_log_file_size 8M 512M innodb_buffer_pool_size 的 25% ，官方推荐是 innodb_buffer_pool_size 的 40-50%, 设置大 一点来避免在日志文件覆写上不必要的缓冲池刷新 行为 innodb_log_buffer_size 128K 64M 用来缓冲日志数据的缓冲区的大小 . 推荐是 8M ， 官方推荐该值小于 16M ，最好是 1M-8M 之间  设计合理的数据表结构：适当的数据冗余 设计合理的数据表结构：适当的数据冗余 执行情况，是否锁表，查看相应的 SQL SQL 语句 语句  设置 设置 my.cnf my.cnf 中的 中的 long-query-time long-query-time 和 和 log-slow-queries log-slow-queries 能 能 够 够 记录服务器那些 记录服务器那些 SQL SQL 执行速度比较慢 执行速度比较慢  另外有用的几个查询： 另外有用的几个查询：

d/authkeys auth 1 1 crc  ha.cf 的配置 master 的 ha.cf 的配置 vim /etc/ha.d/ha.cf logfile /var/log/ha-log logfacility local0 keepalive 2 deadtime 30 warntime 10 initdead 60 udpport 694 ucast hacluster /usr/lib64/heartbeat/ipfail backup 的 ha.cf 的配置 vim /etc/ha.d/ha.cf logfile /var/log/ha-log logfacility local0 keepalive 2 deadtime 30 warntime 10 initdead 60 udpport 694 ucast auth 1 1 crc  ha.cf 的配置 master(dbserver1)的 ha.cf 的配置 vim /etc/ha.d/ha.cf logfile /var/log/ha-log logfacility local0 keepalive 2 deadtime 30 warntime 10 initdead 60 udpport 694 ucast

master_host='master_db', master_user='sync', master_password='sync', master_port=3306, master_log_file='<主数据库查询得到的 File 值>', master_log_pos=<主数据库查询得到的 Positions 值>; 下面是我的脚本例子： change master to master_host='master_db' ', master_user='sync', master_password='sync', master_port=3306, master_log_file='mariadb-bin.000004', master_log_pos=789; /*开启从数据库复制*/ start slave; 最后可以通过 show slave status; 查看同步情况。 至此我们就建立了一个基于 Docker

B+树范围查询 TokuDB事务（一） Page.1 Page.2 Page.3 Memory Disk Redo log (log**.toku) Page.8 Page.9 Page.2 Undo log (tokudb.rollback) FractralTreeIndexFile(*.tokudb) LRU-Cache TokuDB事务（二）

Check MySQL status const String & check_query = "SHOW VARIABLES WHERE " "(Variable_name = 'log_bin' AND upper(Value) = 'ON') " "OR (Variable_name = 'binlog_format' AND upper(Value) = std::stringstream ss; client.getPosition().dump(ss); return ss.str(); }; LOG_INFO(log, "MySQL executed position: \n {}", position_message()); } Consume new data F.A.Q • MySQL

：使skip-slave-start 成为全局、持久、只读的系统变量。 8.0.26 ：组复制成员操作（主成员上可配置super_read_only） 8.0.26：指定用于记录View_change_log_event 的UUID 8.0.27：异步复制通道配置自动跟随主成员 Copyright @ 2021 Oracle and/or its affiliates. MySQL InnoDB root@localhost:3331 --user=mysqlrouter $ sudo systemctl start mysqlrouter $ sudo tail -f /var/log/mysqlrouter/mysqlrouter.log 29 / 55 Copyright @ 2021 Oracle and/or its affiliates. 更改路由器配置选项 更改target_cluster：