Amazon releases a cloud service-based MySQL database

With the emergence of cloud computing, cloud services, and other concepts, databases on the cloud seem to be nothing new. Here we will introduce Amazon's new Amazon RDS, that is, the MySQL database as a cloud service. BKJIA recommends MySQL database entry and mastery tutorial

Amazon recently added a new MySQL database to their Amazon Web Services (AWS) platform called Amazon Relational Database Service (RDS), which can work like a traditional MySQL system. Before RDS, the customer had several options on AWS Database Service:

Customers running on Amazon Machine Image (AMI) provide their own database services

SimpleDB service owned by Amazon Web service

SimpleDB is a simple data storage system. It lacks comprehensive functions of a fully mature Relational Database Management System (RDBMS), but provides a scalable key-value storage. The customer's self-provided database services are not much inferior to the traditional data center environment. The customer's employees are responsible for managing database applications, including configuration, performance tuning, capacity management, and version upgrade, patching and data backup. You can use an interactive tool connected to a traditional MySQL database to control it in the same way.

Amazon RDS allows the customer staff to reduce a lot of MySQL O & M tasks. With it, the scalability and performance monitoring of database computing resources do not require human intervention. Database software is usually patched and backed up by service providers, and is made by the retention period defined by the customer. Scalability comes from AWS's so-called "instance classes", which have a total of five. You can gradually increase from a common virtual CPU kernel and a memory of GB to a super large database instance ", that is, 68 Gb memory and 8 virtual CPU cores. After 100% of the Active Database Data occupies the backup storage, additional storage space is charged. In addition, data exists in another zone rather than where the instance is located. This is similar to the concept of remote data protection in the traditional data security model.

Thanks to flexibility, AWS defines a 4-hour maintenance window every week. This maintenance window can be used to patch the application software and back up data. The customer cannot choose to exit the patching process. However, they can specify when the maintenance window will occur within one week. In the maintenance window, the database instance is offline for a specific period of time. Amazon states that "in rare cases, patching takes longer than part of your maintenance window, even if it only happens for security or persistence-related patches ."

This means that the customer must anticipate and plan such a weekly offline instance event. Even if the service provider says it is unlikely that four hours will be used up, the customer will expect the worst case, with four hours of offline instance time per week. For customers who can accept a relatively short period of time for database instances to be unavailable, the planned shutdown time may be acceptable, with only the minimum possible impact. However, some customers do not have the freedom to make such choices. They must ensure that the service is available 24x7, even when the maintenance window is run every week. In traditional database deployment, database replication technology is often used to achieve high availability. Can the replication technology also be used in RDS to allow customers to specify different maintenance opportunities for different database instances? For example, are the following situations possible?

◆ Are two or more instances running in master-slave mode?

◆ Are two instances running in master-master mode?

◆ Are two or more instances running in cluster mode?

There is no clear answer yet. In the "new features coming soon" section on the RDS service details page, Amazon's expected data replication availability options will be:

High Availability-developers and business users who want to exceed the flexibility beyond Amazon RDS's automatic backup will not have to pay extra for this. With high availability, they can easily replicate database instances between multiple zones at cost effectiveness to prevent failures caused by a single storage.

It seems that this will solve the availability problem at the cost of multiple zones. Traditional technologies such as master-slave and master-master models that address availability cannot play a role in this regard.

1. Initial Experience of MySQL and Amazon RDS in the cloud
2. Top 10 MySQL optimization skills for WEB programmers
3. What Will MySQL do if it is abandoned by Oracle?
4. MySQL left connection, right connection, and internal connection
5. Founder of PostgreSQL: MySQL recession is inevitable