Database server
16 cores
63Gb RAM
CentOS release 6.8
etc. /my.cnf
[mysqld]
pid_file=/var/lib/mysql/fatty01.pid
datadir=/var/lib/mysql
socket=/var/lib/mysql/mysql.sock
user=mysql
innodb_buffer_pool_size = 50G
innodb_log_file_size = 2G
innodb_flush_log_at_trx_commit = 0
sync_binlog = 0
innodb_flush_method = O_DIRECT
innodb_buffer_pool_instances = 16
innodb_thread_concurrency = 16
skip_name_resolve = 1
innodb_io_capacity = 4000
innodb_io_capacity_max = 6000
innodb_buffer_pool_dump_at_shutdown = 1
innodb_buffer_pool_load_at_startup = 1
query_cache_size = 0
query_cache_type = OFF
innodb_checksum_algorithm = crc32
table_open_cache_instances = 16
innodb_read_io_threads = 20
innodb_write_io_threads = 10
max_connections = 700
when we have peaks from 3000 simultaneous clients, mysqld does not seem to pull all resources from the machine.
I see a load of 40, but the processor does not seem to overcome 60%
This is reflected on the external server.
** My question is clear, how can I improve performance without compromising the server? You can also reduce the MYSQL wait time on an external server, since there is clearly a problem with the configurations on the server side of the database. **
**
UPDATE After investigation, the problem seems to be related to slow queries, so I’ll assume that this configuration is optimal for this equipment
**
source
share