The case of 12306 and Hadoop tells us that the root causes of NOSQL technology and the main application scenarios:
(1), to solve the data storage and access problems that cannot be solved by traditional relational databases
(2), to solve the big data application problems
(3), To solve the application problem on the Internet (concurrency problem)
1. Nosql definition: The
main body conforms to the non-relational, distributed, open source, and horizontally scalable next-generation database. The
English name NOSQL itself means "Not Only SQL", which means "Not just SQL"
Horizontal expansion: Cluster
Vertical expansion: Machine performance
2. Time units
Some time units often used in NOSQL are: seconds (s), milliseconds (ms), microseconds (μs), nanoseconds (ns), the relationship between them is 1 s = 10^3 ms = 10^6 us = 10^9 ns.
3. Storage data unit
4. Logical mode definition,
also called conceptual mode, is the description of the logical structure and characteristics of all data in the database, and is the common data view for all users.
5. Big Data definition 4v+1O
Gartner defines big data as high-speed , Huge and (or) changeable data. The so-called high-speed refers to the rapid generation or change of data. The so-called huge amount refers to the large scale of data. The so-called changeable refers to the range of data types or the range of information contained in the data is very wide
6. Hash function (HashFuncation)
business primary key: such as student number, the code itself is the only
logical primary key: 1, 2, 3, 4
The hash function, also known as the hash function, is an algorithm that can map the input value to the output string.
The hash value generated by the hash function is unique (in rare cases, collision problems may occur).
For example, Hash ("name: Zhang Hua") will generate a hash value (hexadecimal number) similar to 4b28c82d91e028181f2128392ab9219e.
7. Database classification
(1) The
traditional relational database technology of trdb was originally designed based on the concept of centralized management of data on a single machine, so it is limited by the physical performance of a single machine (usually a server).
Traditional relational database technologies are all built on a centralized database management system (DBMS, Database Management System),
which is to process data through a relational database system. Under normal circumstances, it can only be implemented in a single machine, and horizontally multi-machine expansion (distributed application) There are great difficulties
(2). Nosql has
given birth to the rapid development of NOSQL technology due to the emergence of big data problems. It has to make up for the technical shortcomings of traditional relational databases-especially in the processing of speed, storage capacity and diversified structure data. On the issue.
Generally, NOSQL technology can be divided into five modes: key value storage, document storage, column family storage, graph storage, and other storage based on the principle of data storage structure.
(3) NewSQL database
The purpose of its appearance is to combine the advantages of traditional relational database and NOSQL database technology to realize data storage and processing in a big data environment.
From the above sentence, it can be seen that NewSQL database designers must not only realize the fast and effective big data processing capabilities of NOSQL technology, but also realize the advantages of SQL and transaction processing of traditional relational databases. I kind of want to eat both.