Analyze MySQL performance optimization points from large to small, from primary to secondary, to achieve the ultimate optimization.
Using Mindmanger to organize ideas, such as, each point can be found on the internet to explain, and recorded. The idea of optimization was formed:
1 Connecting Connections
Often meet "mysql:error 1040:too many connections" situation, one is the traffic is really high, MySQL server can not resist, this time to consider increasing the load from the server to read the pressure, another situation is the MySQL configuration file max_ Connections value is too small:
Mysql> Show variables like ' max_connections ';
+-----------------+-------+
| variable_name | Value |
+-----------------+-------+
| max_connections | 256 |
+-----------------+-------+
The maximum number of connections for this MySQL server is 256, then query the maximum number of connections the server responds to:
Mysql> show global status like ' Max_used_connections ';
MySQL server in the past the maximum number of connections is 245, did not reach the maximum number of server connections 256, should not appear 1040 error, the more ideal setting is
Max_used_connections/max_connections * 100%≈85%
The maximum number of connections is about 85% of the maximum number of connections, if the ratio is found below 10%, the maximum number of MySQL server connections is set too high.
2 Threads Thread
Mysql> show global status like ' thread% ';
+-------------------+-------+
| variable_name | Value |
+-------------------+-------+
| threads_cached | 46 |
| threads_connected | 2 |
| threads_created | 570 |
| threads_running | 1 |
+-------------------+-------+
If we set the thread_cache_size in the MySQL server configuration file, when the client disconnects, the server processes the client's thread to cache in response to the next customer instead of destroying it (provided the cache count is not up to the limit).
Threads_created indicates the number of threads created, if the threads_created value is found to be too large, it indicates that the MySQL server has been creating threads, which is also a relatively resource-intensive, can appropriately increase the thread_cache_size value in the configuration file,
Query Server Thread_cache_size configuration:
Mysql> Show variables like ' thread_cache_size ';
+-------------------+-------+
| variable_name | Value |
+-------------------+-------+
| Thread_cache_size | 64 |
+-------------------+-------+
The server in the example is still quite healthy.
3 Caching Cache
3.1 File opening numbers
Mysql> show global status like ' Open_files ';
+---------------+-------+
| variable_name | Value |
+---------------+-------+
| Open_files | 1410 |
+---------------+-------+
Mysql> Show variables like ' open_files_limit ';
+------------------+-------+
| variable_name | Value |
+------------------+-------+
| Open_files_limit | 4590 |
+------------------+-------+
More appropriate setting: Open_files/open_files_limit * 100% <= 75%
3.2 Data Sheets
3.2.1 Open Number Open_tables
Mysql> show global status like ' open%tables% ';
+---------------+-------+
| variable_name | Value |
+---------------+-------+
| Open_tables | 919 |
| Opened_tables | 1951 |
+---------------+-------+
Open_tables: Number of open tables
Opened_tables: Number of open tables
If the number of opened_tables is too large, the value of Table_cache (which is called Table_open_cache after 5.1.3) in the configuration may be too small, let's check the server Table_cache value:
Mysql> Show variables like ' Table_cache ';
+---------------+-------+
| variable_name | Value |
+---------------+-------+
| Table_cache | 2048 |
+---------------+-------+
The more appropriate values are:
Open_tables/opened_tables * 100% >= 85%
Open_tables/table_cache * 100% <= 95%
3.2.2 Temporary table tmp_table
Mysql> show global status like ' created_tmp% ';
+-------------------------+---------+
| variable_name | Value |
+-------------------------+---------+
| Created_tmp_disk_tables | 21197 |
| Created_tmp_files | 58 |
| Created_tmp_tables | 1771587 |
+-------------------------+---------+
Each time a temporary table is created, created_tmp_tables increases, if the temporary table is created on disk, Created_tmp_disk_tables also increases, created_tmp_files represents the number of temporary file files created by the MySQL service, The more ideal configuration is:
Created_tmp_disk_tables/created_tmp_tables * 100% <= 25%
For example, the server above created_tmp_disk_tables/created_tmp_tables * 100% = 1.2%, should be quite good. Let's look at the MySQL server configuration for the staging table:
Mysql> Show variables where variable_name in (' Tmp_table_size ', ' max_heap_table_size ');
+---------------------+-----------+
| variable_name | Value |
+---------------------+-----------+
| Max_heap_table_size | 268435456 |
| Tmp_table_size | 536870912 |
+---------------------+-----------+
Only temporary tables below 256MB can be used for all memory, and the hard disk temporary tables will be available.
3.2.3 Table Lock Condition
Mysql> show global status like ' table_locks% ';
+-----------------------+-----------+
| variable_name | Value |
+-----------------------+-----------+
| Table_locks_immediate | 490206328 |
| table_locks_waited | 2084912 |
+-----------------------+-----------+
Table_locks_immediate indicates that the number of table locks is released immediately,
Table_locks_waited indicates the number of table locks that need to wait.
If the table_locks_immediate/table_locks_waited > 5000, it is best to use the InnoDB engine, because InnoDB is a row lock and MyISAM is a table lock, for high concurrent write application InnoDB effect will be better.
The server table_locks_immediate/table_locks_waited = 235,myisam In the example is sufficient.
3.2.4 Table Scan Condition
Mysql> show global status like ' handler_read% ';
+-----------------------+-------------+
| variable_name | Value |
+-----------------------+-------------+
| Handler_read_first | 5803750 |
| Handler_read_key | 6049319850 |
| Handler_read_next | 94440908210 |
| Handler_read_prev | 34822001724 |
| Handler_read_rnd | 405482605 |
| Handler_read_rnd_next | 18912877839 |
+-----------------------+-------------+
For each field explanation see http://hi.baidu.com/thinkinginlamp/blog/item/31690cd7c4bc5cdaa144df9c.html, call up the number of query requests completed by the server:
Mysql> show global status like ' Com_select ';
+---------------+-----------+
| variable_name | Value |
+---------------+-----------+
| Com_select | 222693559 |
+---------------+-----------+
Calculate table Scan Rate:
Table Scan rate = Handler_read_rnd_next/com_select
If the table scan rate exceeds 4000, indicating that there are too many table scans, it is likely that the index is not built, and that increasing the read_buffer_size value will have some benefits, but it is best not to exceed 8MB.
3.3 Key_buffer_size
Key_buffer_size is one of the most influential parameters for MyISAM table performance, and the following is a configuration of the primary storage engine server with MyISAM:
Mysql> Show variables like ' key_buffer_size ';
+-----------------+------------+
| variable_name | Value |
+-----------------+------------+
| Key_buffer_size | 536870912 |
+-----------------+------------+
Allocated 512MB of memory to Key_buffer_size, let's look at the usage of key_buffer_size:
Mysql> show global status like ' key_read% ';
+------------------------+-------------+
| variable_name | Value |
+------------------------+-------------+
| key_read_requests | 27813678764 |
| Key_reads | 6798830 |
+------------------------+-------------+
A total of 27,813,678,764 index read requests, with 6,798,830 requests not found in memory directly from the hard disk to read the index, calculate the probability of index misses cache:
Key_cache_miss_rate = key_reads/key_read_requests * 100%
For example, the above data, Key_cache_miss_rate is 0.0244%, 4,000 index read requests have a direct read hard disk, already very bt, key_cache_miss_rate in 0.1% The following are good (every 1000 requests have a direct read hard disk), if the key_cache_miss_rate under 0.01%, key_buffer_size allocation of too much, can be appropriately reduced.
Note The default value of Key_read_buffer is 8M. On a proprietary database server, this value can be set to RAM * 1/4
The MySQL server also provides the key_blocks_* parameter:
Mysql> show global status like ' key_blocks_u% ';
+------------------------+-------------+
| variable_name | Value |
+------------------------+-------------+
| key_blocks_unused | 0 |
| key_blocks_used | 413543 |
+------------------------+-------------+
Key_blocks_unused indicates the number of unused cache clusters (blocks)
Key_blocks_used indicates the largest number of blocks ever used
such as this server, all the cache is used, either increase the key_buffer_size, or the transition index, the cache is full. More Ideal settings:
Key_blocks_used/(key_blocks_unused + key_blocks_used) * 100%≈80%
3.4 Sort Usage Sort_buffer
Mysql> show global status like ' sort% ';
+-------------------+------------+
| variable_name | Value |
+-------------------+------------+
| sort_merge_passes | 29 |
| Sort_range | 37432840 |
| Sort_rows | 9178691532 |
| Sort_scan | 1860569 |
+-------------------+------------+
The sort_merge_passes consists of two steps. MySQL first tries to sort in memory, the memory size is determined by the system variable sort_buffer_size, if it is not large enough to read all the records into memory, MySQL will save the results of each in-memory sorted into a temporary file, and so on MySQL Once all the records have been found, the records in the temporary file are sorted once. This re-ordering will increase the sort_merge_passes. In fact, MySQL will use another temporary file to save the re-ordering results, so you will usually see that the sort_merge_passes increment is twice times the number of temporary files built. Because temporary files are used, the speed may be slower, and increasing sort_buffer_size will reduce the number of sort_merge_passes and temporary files created. But blindly increasing sort_buffer_size does not necessarily improve speed, see how fast can you sort data with MySQL? (quoted from http://qroom.blogspot.com/2007/09/ Mysql-select-sort.html, seemingly by the wall)
In addition, increasing the value of Read_rnd_buffer_size (3.2.3 is Record_rnd_buffer_size) is also a bit of a benefit for sorting operations, see: http://www.mysqlperformanceblog.com/ 2007/07/24/what-exactly-is-read_rnd_buffer_size/
3.5 Query Cache
Mysql> show global status like ' qcache% ';
+-------------------------+-----------+
| variable_name | Value |
+-------------------------+-----------+
| Qcache_free_blocks | 22756 |
| Qcache_free_memory | 76764704 |
| Qcache_hits | 213028692 |
| Qcache_inserts | 208894227 |
| Qcache_lowmem_prunes | 4010916 |
| qcache_not_cached | 13385031 |
| Qcache_queries_in_cache | 43560 |
| Qcache_total_blocks | 111212 |
+-------------------------+-----------+
MySQL Query cache variable interpretation:
Qcache_free_blocks: The number of contiguous memory blocks in the cache. A large number indicates that there may be fragmentation. Flush query Cache organizes the fragments in the cache to get a free block.
Qcache_free_memory: Free memory in the cache.
Qcache_hits: Increases each time the query hits the cache
Qcache_inserts: Increases each time a query is inserted. The hit count divided by the number of inserts is the hit ratio.
Qcache_lowmem_prunes: The number of times that the cache is out of memory and must be cleaned up to provide space for more queries. This number is best seen over a long period of time, and if the number is growing, it can mean that fragmentation is very serious, or that there is little memory. (The above free_blocks and free_memory can tell you which kind of situation)
Qcache_not_cached: The number of queries that are not appropriate for caching, usually because these queries are not a SELECT statement or are using functions such as now ().
Qcache_queries_in_cache: The number of queries (and responses) that are currently cached.
Qcache_total_blocks: The number of blocks in the cache.
Let's look at the server's configuration for Query_cache:
Mysql> Show variables like ' query_cache% ';
+------------------------------+-----------+
| variable_name | Value |
+------------------------------+-----------+
| Query_cache_limit | 2097152 |
| Query_cache_min_res_unit | 4096 |
| Query_cache_size | 203423744 |
| Query_cache_type | On |
| Query_cache_wlock_invalidate | Off |
+------------------------------+----------+
Explanation of each field:
Query_cache_limit: Queries that exceed this size will not be cached
Query_cache_min_res_unit: Minimum size of cache block
Query_cache_size: Query Cache Size
Query_cache_type: Cache type, determines what queries are cached, example indicates that select Sql_no_cache queries are not cached
Query_cache_wlock_invalidate: When there are other clients that are writing to the MyISAM table, if the query is to be returned with the cache result or wait for the write operation to complete, the table gets the result.
Query_cache_min_res_unit configuration is a "double-edged sword", the default is 4KB, set the value of large data query is good, but if your query is small data query, it is easy to create memory fragmentation and waste.
Query Cache Fragmentation Rate = Qcache_free_blocks/qcache_total_blocks * 100%
If the query cache fragmentation rate exceeds 20%, you can use flush query cache to defragment the cache, or try to reduce query_cache_min_res_unit if your queries are small data volumes.
Query Cache utilization = (query_cache_size-qcache_free_memory)/query_cache_size * 100%
Query cache utilization below 25% indicates that the query_cache_size setting is too large to be appropriately reduced; query cache utilization is above 80% and Qcache_lowmem_prunes > 50 query_cache_ Size may be a little bit small, or too much fragmentation.
Query Cache Hit Ratio = (qcache_hits-qcache_inserts)/qcache_hits * 100%
Sample server query Cache Fragmentation rate = 20.46%, query cache utilization = 62.26%, query cache Hit Ratio = 1.94%, hit ratio is poor, probably write more frequently, and possibly some fragments.
4 other
4.1 read_buffer_size
4.2 Slow Query
Mysql> Show variables like '%slow% ';
+------------------+-------+
| variable_name | Value |
+------------------+-------+
| log_slow_queries | On |
| Slow_launch_time | 2 |
+------------------+-------+
Mysql> show global status like '%slow% ';
+---------------------+-------+
| variable_name | Value |
+---------------------+-------+
| Slow_launch_threads | 0 |
| slow_queries | 4148 |
+---------------------+-------+
The configuration of the record slow query, the execution time of more than 2 seconds is slow query, the system shows that there are 4,148 slow query, you can analyze the slow query log, find the problem of SQL statements, slow query time should not be set too long, otherwise the meaning is not very good, preferably within 5 seconds, if you need a microsecond level of slow query, Consider a patch for MySQL: Http://www.percona.com/docs/wiki/release:start, remember to find the corresponding version.
Turning on the slow query log may have a little impact on system performance, and if your MySQL is a master-slave structure, you can consider opening one of the slow query logs from the server, which can monitor slow queries and have little impact on system performance.
Full instructions and file download
MySQL Performance tuning direction
Start building with 50+ products and up to 12 months usage for Elastic Compute Service
1 on 1 presale consultation
24/7 Technical Support 6 Free Tickets per Quarter Faster Response
Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.
A comprehensive suite of global cloud computing services to power your business
Payment Methods We Support
