Database operation is difficult. There are many things to consider such as load balancing, backup, patch application. It is even harder when a failure occurs. It will be busy to replace hardware or recover from backup. Amazon RDS is a distributed relational database service of Amazon Web Services (AWS). It is a service that provides a relational database as ‘Amazon Relational Database Service.
RDS is a web service that operates in the cloud designed to simplify the configuration, operation, and scaling of relational databases used in applications. By activating various RDBMS such as MySQL, MS SQL Server, Oracle etc on AWS, users can use database service without hardware.
In some RDBMSs, it is also possible to make redundant configuration across Availability Zone with one click. However, since it is about database service, users can not log in to the operating system where RDS is running. Complex management processes such as database software patch, database backup, point-in-time recovery etc are automatically managed. Storage scaling and resource computation can be done with a single API call.
Benefits of Using Amazon RDS
Available Immediately – Installation of OS and database is unnecessary, so you can start using it in a few minutes. Just specify the type and size of the database from the AWS console. Alternatively, it can be created from a command line tool or API.
No maintenance required – Regularly patched patches are applied automatically. Security is always enhanced by always applying the latest patch. The patch is applied within the specified maintenance window time. In this case, downtime occurs, but by setting the multi AZ arrangement, downtime can be minimized.
Dedicated Monitoring Service – An instance of RDS can be monitored using Amazon Cloud Watch. You can monitor CPU usage, memory usage, network usage, etc. If an alarm is set, it can be automatically scaled when the threshold is exceeded.
Can be Scaled Up / Down Anytime – By changing the instance class, you can easily scale up CPU power, memory size, and network performance. Downtime will occur for several minutes, but you can change it at any time. Also, you can change the database size without downtime. Automatic scale setting is also possible.
Backup Support – In RDS, automatic backup is enabled by default. You can keep the backup for up to 35 days and restore it in case of emergency. Also, using snapshots, you can save databases at any time. You can reproduce exactly the same database by creating a new instance from the snapshot.
Availability with Amazon RDS
In RDS, there are two types of Single availability zones (AZ) placement and Multi AZ placement, and if you select Multi AZ placement, the real and standby instances are automatically placed in multiple availability zones. Data synchronization between the production system (master) and the standby system (slave) is done automatically, and in the unlikely event you switch from the master to the slave, the system stop can be minimized. This is a way to realize a mechanism with high availability.
How Does It Work?
Amazon RDS is written as a fully managed type, a relational database service that is easy to set up, operate, and extend. It is fully managed and easy to operate.
Database Creation – One can create a database with a few clicks. Just select the database engine, instance class, type and size of the disk etc.
Lengthy – Redundant configuration can be constructed between data centers called Multi AZ with one click. It is always synchronously replicated, and it automatically fails over when a failure occurs. Manual forced failover is also possible.
Backup – It automatically acquires snapshots on a daily basis and saves the transaction log every five minutes. You can create a database instance from the snapshot or create a database instance with the time specified by Point – In – Time – Recovery (PITR). It is possible to back up at any time by manual operation.
Apply Patch – If patch application is required due to database vulnerability etc., it will be applied automatically. By setting the maintenance window, you can specify the patch application timing. Because it performs a rolling update if the multi AZ configuration is done, the service stop time becomes short.