Buffer pool related
select @@innodb_buffer_pool_size;
+---------------------------+
| @@innodb_buffer_pool_size |
+---------------------------+
| 1073741824 |
+---------------------------+
select 1073741824/1024/1024;
+----------------------+
| 1073741824/1024/1024 |
+----------------------+
| 1024.00000000 |
+----------------------+
# innodb_buffer_pool_size 一般建议最多是物理内存的 75-80%the innodb_flush_log_at_trx_commit (one a double standard)
The main control innodb data log buffer is written to the log file to disk and flush points, respectively 0,1,2 three values.
select @@innodb_flush_log_at_trx_commit;
+----------------------------------+
| @@innodb_flush_log_at_trx_commit |
+----------------------------------+
| 1 |
+----------------------------------+
参数说明:
1，每次事物的提交都会引起日志文件写入、flush 磁盘的操作，确保了事务的ACID；flush 到操作系统的文件系统缓存 fsync到物理磁盘.
0，表示当事务提交时，不做日志写入操作，而是每秒钟将 log buffer 中的数据写入文件系统缓存并且秒 fsync 磁盘一次；
2，每次事务提交引起写入文件系统缓存,但每秒钟完成一次fsync磁盘操作。
--------
The default setting of 1 is required for full ACID compliance. Logs are written and flushed to disk at each transaction commit.
With a setting of 0, logs are written and flushed to disk once per second. Transactions for which logs have not been flushed can be lost in a crash.
With a setting of 2, logs are written after each transaction commit and flushed to disk once per second. Transactions for which logs have not been flushed can be lost in a crash.
-------Innodb_flush_method=(O_DIRECT, fdatasync)
Can refer to the official document: https: //dev.mysql.com/doc/refman/5.7/en/innodb-parameters.html#sysvar_innodb_flush_method
This option controls the log buffer and the data buffer is flashed when the disk through the file system cache whether
Check the status quo strategy
show variables like '%innodb_flush%';
# 参数说明
O_DIRECT : 数据缓冲区写磁盘,不走 OS buffer
fsync : 日志和数据缓冲区写磁盘,都走 OS buffer
O_DSYNC : 日志缓冲区写磁盘,不走 OS bufferRecommendations
最高安全模式:
innodb_flush_log_at_trx_commit=1
Innodb_flush_method=O_DIRECT
最高性能:
innodb_flush_log_at_trx_commit=0
Innodb_flush_method=fsync