MySQL Database Overview
The following features are found in MySQL database. It is possible to operate smoothly and conduct installation and settings in a short amount of time. The database is superior regarding the functions of replication and partition, and it is also excellent with availability, management, and safety. It has a transaction function and keeps high reliability of data. The stored procedure “trigger view” supports information protection and the development and operation of systems. With database management, MySQL’s database is superior in operability, so it is possible to implement smoothly. Also, as it can automatically manage the functions of the cluster database etc., you don’t have to spend as much in labor.
Replication is copying data from a database in one server to a database in another.
Its purposes include the distribution of server loads, prevention of system-down, and operative continuation during failures by using master and slave.
Since replication is executed over LAN or internet, it is also possible to replicate data to the server that is geographically far from the master database. MySQL has different types of replication as below.
|Row-based||Send SQL results (rows) run by the master to the slave, making the data consistent.|
|Statement-based||Send SQL run by the master to the slave and run it again, making the data consistent.|
|Mixed-based||Switch the methods above in logging events|
MySQL Replication Moniter monitors the status of replication. By using the replication adviser rules, it is possible to detect potential problems before they actually impact replication. After the detection, the replication can be stabilized by addressing such potential problems.
Partitioning means to divide database rows into small data set and preserve them in directories and disks.
In MySQL, horizontal partitioning is supported.
Thanks to this, it is possible to access only the specified partition.
As a result, the query speed and database performance are improved.
If you want to use partitioning, you only have to specify partitioning from the options on the initial creation, and you no longer have to worry about it while extracting data.
These are the following types of partitioning: RANGE, LIST, COLUMN, HASH, KEY, and SUB.
|RANGE||Method to set the value range of the partition. For example, this method is used when you divide data into generations. It is easy to check the data for every partition, but the deviation occurs in the data amount.|
|LIST||Form to set values instead of ranges of partitions. It is easy to check the data of each partition, but the deviation occurs in the amount of data.|
|COLUMN||Form to designate plural rows to partitioning keys. This is the expanded RANGE and LIST.|
|HASH||Form to calculate the HASH values from 1 row, or plural rows, and assign to partitions, based on the values. It is possible to assign the amount of data to partitions equally.|
|KEY||Use this in case you want to designate character string types and plural keys. Partition by the HASH value calculated by the system.|
|Partition Pruning||This divides the content of the partitioned table into further pieces. Use this in case of mass tables.|
|SUB||This is the function to access (refer to) only the particular partition. Compared with the case of not being partitioned, the processing speed becomes fast as it specializes and conducts.|