Is NoSQL true or not?

With the emergence of a large amount of data, NoSQL database architecture is being used by more and more people, because NoSQL architecture is more effective in big data processing. Today, when you need to save

With the emergence of a large amount of data, NoSQL database architecture is being used by more and more people, because NoSQL architecture is more effective in big data processing. Today, when you need to save

To NoSQL or not toNoSQL

With the emergence of a large amount of data, NoSQL database architecture is being used by more and more people, because NoSQL architecture is more effective in big data processing.

Today, when you need to store and process high-capacity data, many young developers prefer open-source NoSQL. They claim that NoSQL is the database architecture of the future, and even some Silicon Valley startups believe that traditional relational databases have no significance.

At the same time, experienced DBAs and developers put forward an example worth noting: Due to the NoSQL vulnerabilities, hackers can use these vulnerabilities to successfully access the system, this led to the failure of the Flexcoin and Poloniex Bitcoin trading companies. Even those who stick to RDBMS believe that NoSQL is not suitable for the development of any key application (missioncritical.

Of course, the truth is certainly in the middle of this two-pole viewpoint. Both architectures have their unique characteristics and suitable scenarios, so they should coexist. It is more useful to focus on the data accuracy required by the actual application, especially when inconsistent data encounters a low-performance environment. E-commerce, finance, shipping, logistics, and manufacturing are all like this. For these industries, "eventual consistency" cannot fully meet their needs, inconsistent databases may make the business worse.

For key task development, administrators need immediate consistency instead of "final consistency" to ensure that data can be used immediately and consistent in applications ,, you do not need to wait until the application operation is complete. To achieve this level of security protection, developers need the database to have ACID features for each transaction. Eventual consistency cannot meet the needs of these industries. Only a complete ACID database can be solved. Because of the ACID characteristics, the operations of each transaction are reliable and consistent, at the same time, ACID features ensure data consistency and will not be affected by operation accidents or external interference.

At the same time, some popular open-source NoSQL databases, such as MongoDB and Cassandra, make it easier to build and quickly write and deploy applications. Their speed has also attracted many companies that are scrambling to seize the market. Today, these databases also propose a new concept BASE corresponding to ACID (basic availability, soft state, and final consistency ).

Although the popular NoSQL database architecture does not support ACID, some NoSQL integration implementations are acceptable. These integration implementations can even achieve the immediate consistency ACID features required by key tasks. In fact, some financial service enterprises are using NoSQL databases with real-time consistency to support millions of financial transactions per second in a strictly managed environment.

There are many online retailers who need immediate and consistent ACID support, but they also need to be able to quickly process data such as user feedback, this data does not require strong ACID support, so the final consistency can meet the processing needs.

For anyone who is determined to use NoSQL or SQL, they should realize that SQL architecture does have a suitable scenario in financial applications, such as analyzing a report in a standard format.

Therefore, for key task applications that require a short period of time to process a large amount of data and expand elasticity, real-time accuracy is also required, then they should choose NoSQL databases with instant consistency ACID support.

For scenarios that do not require real-time data consistency, such as social network content and other scenarios that only need to ensure eventual consistency, the NoSQL supporting BASE is enough to meet the requirements.

Traditional SQL is the most competent for scenarios that require deep analysis, reporting, and other strict business operations.

Finally, there are a lot of database customers to choose from, and even some can provide the same dataset NoSQL and SQL mixed use technology. In the face of these various choices, we still need to consider the actual application scenarios and make the choices. SQL and NoSQL have their own advantages and disadvantages, and only use them in the most suitable scenarios, in order to play their biggest role!

Quick Start to NoSQL databases. For details about how to download high-definition PDF, see

Basic knowledge about NoSQL Databases

Key to enterprise application of NoSQL