Cloud storage is not the terminator of a relational database

 

Prior to the advent of cloud computing, relational databases such as Oracle, MSSQL, MySQL, and Postgres have dominated the data kingdom, and few have been able to shake it. But now the age of information has changed this situation, with the rapid growth of data access, the shortcomings of relational database extensibility gradually exposed. Although it is possible to solve the problem of insufficient expansion in the short term by increasing the memory and updating the data management system, this does not conform to the sustainable development rule.
Cloud storage technology from cloud computing was first invented in the Internet enterprise. Cloud storage technology is not specific to a technology, but a large class of technology collectively, it through the cluster application, grid technology or distributed file system or similar grid computing and other functions to work together, and with a certain application software or application interface, to provide users with certain types of storage services and access services. Complex back-end support enables cloud storage to have the advantage of having more relational databases on the front-end: Easy to scale, reliable and secure, resource controllable, high utilization, low cost, and fast speed.
These advantages enable cloud storage technology to have its own areas of expertise, such as statistical analysis of data daily traffic is very large, the need for cloud storage such a large-scale distributed deployment. But in the face of very high consistency requirements of transactional intensive computing, cloud storage seems to be inadequate. In general, there are three major requirements for a large-scale distributed database system, namely consistency, availability, and partition tolerance, where traditional relational databases have limited support for partition tolerance, but consistency and availability are their advantages. In other words, the short board of cloud storage is the advantage of relational database, and they are complementary to some extent.
On the one hand, the current domestic already has Jinshan fast plate, 115 network disk, Baidu network disk occupy the market, and are in the massive free marketing. On the other hand, there are some strong supporters of relational databases in traditional enterprises. In this context, users can easily get lost in their eyes and choose a storage service that is not suitable for them. Cloud storage is not a relational database terminator, at present can not jump to the conclusion, as a user, we need to throw away the exciting advertising, calmly analyze their own needs. In the sense of responsibility, it is incumbent on the merchant to help customers choose a database that suits their application, maximizing efficiency and minimizing costs.

Cloud storage is not the terminator of a relational database