Analyzing the MySQL Slow Query Log to Find Performance Bottlenecks

Turn on MySQL slow query logs and analyze locationable performance issues. 1. Edit the configuration file or dynamically set slow_query_log and long_query_time; 2. The log contains key fields such as Query_time, Lock_time, Rows_examined to assist in judging efficiency bottlenecks; 3. Use mysqldumpslow or pt-query-digest tools to efficiently analyze logs; 4. Optimization suggestions include adding indexes, avoiding SELECT *, splitting complex queries, etc. For example, adding an index to user_id can significantly reduce the number of scanned rows and improve query efficiency.

MySQL's slow query log is an important tool for troubleshooting performance issues. If you find that the database response is slow, or some page loading time increases significantly, turning on and analyzing slow query logs can often quickly locate the problem.

Turn on slow query log

First, make sure that the slow query log is enabled and a suitable definition of "slow" is set. By default, this value is 1 second, but you can adjust it according to actual needs:

• Edit my.cnf or my.ini file:

   slow_query_log = 1
  slow_query_log_file = /var/log/mysql/mysql-slow.log
  long_query_time = 0.5

It can also be dynamically set through SQL:

 SET GLOBAL slow_query_log = 'ON';
SET GLOBAL long_query_time = 0.5;

Note: After modifying the parameters, you may need to reconnect or refresh the session to take effect.

Analyze slow query log content

Each line of the record in the log file contains execution time, lock time, return number of rows, scan number of rows, and actual execution of SQL statements. for example:

 # Query_time: 2.34 Lock_time: 0.00 Rows_sent: 10 Rows_examined: 100000
SELECT * FROM orders WHERE user_id = 123;

Although the above SQL only returns 10 data, it scans 100,000 rows, indicating that it is likely that the index is missing or the query method is not efficient enough.

Several common key fields:

• Query_time : The time taken in the entire query (seconds)
• Lock_time : the time to wait for the lock
• Rows_examined : Number of rows scanned
• Rows_sent : The number of rows sent to the client

If Rows_examined is much larger than Rows_sent , then you need to consider optimizing the index or query structure.

Recommended commonly used analysis tools

Manual log viewing is inefficient, and some tools can be used to help analyze:

• mysqldumpslow : MySQL comes with command line tool, which can count and summarize slow queries.

   mysqldumpslow -s at -t 10 /var/log/mysql/mysql-slow.log

  The above command will sort by average time and list the top 10 slowest queries.

• pt-query-digest : A tool in Percona Toolkit, which is more powerful and supports more complex aggregation and analysis.

   pt-query-digest /var/log/mysql/mysql-slow.log > report.txt

These tools can help you find out which SQLs appear frequently and consume more resources, so as to prioritize optimization.

Common optimization suggestions

Once you find the slow queries, the next step is to optimize them. Here are some common practices:

• Index fields that are frequently queried, especially fields in WHERE and JOIN conditions
• Avoid using SELECT * and select only necessary fields
• Use EXPLAIN to view the execution plan and confirm whether the index is hit
• Reasonably split complex queries to avoid long-term use of resources in large queries
• Do table analysis regularly ( ANALYZE TABLE ) to keep statistical information accurate

For example, the above mentioned SELECT * FROM orders WHERE user_id = 123; if no index is established on user_id , you can greatly improve performance by adding indexes:

 ALTER TABLE orders ADD INDEX idx_user_id (user_id);

When this is executed again, the number of scanned rows may drop from tens of thousands to tens of dozens, which is very obvious.

Basically that's it. It is not particularly difficult to analyze slow query logs, but it is indeed easy to ignore some details, such as whether the index actually works, whether the execution plan has changed, etc. As long as you insist on regular inspections, you can promptly discover potential problems SQL to avoid performance deterioration.

The above is the detailed content of Analyzing the MySQL Slow Query Log to Find Performance Bottlenecks. For more information, please follow other related articles on the PHP Chinese website!