Work often encountered problems, called me a headache, a little to tidy up a plan.
1. Replication latency due to too many slave libraries
Optimization: Recommended from the number of libraries 3-5 advisable (specifically to see their own hardware configuration)
2. From the library hardware is worse than the main library hardware
Optimize: Improve hardware performance
3. Too many slow SQL statements
Optimization: SQL statement execution time is too long to optimize SQL statements (need to contact DBA development to discuss optimization statements)
4. Master-slave Replication design Issues
Optimization: Master-slave replication single-threaded, can be resolved through multi-threaded IO solution, and MySQL5.6.3 support multi-threaded IO replication.
5. Network latency between master and slave libraries
Optimization: Try to link short, increase port bandwidth
6. High reading and writing pressure in the main library
Optimization: Front-end plus buffer and cache. Master-Slave delay is not synchronized (how much delay, as long as not affect the business is OK)
7, Business design defects cause delay impact business
Optimization: No data changed from library to main library
This article is from the "I am a Little bird" blog, please make sure to keep this source http://liyanchao.blog.51cto.com/9954601/1683560
How does MySQL synchronization delay work? Optimization?