Starting from the classification of database, I understand the graph database

stay CSDN Neo4j The skill tree After the release of , There has been a 64 Students are learning , Many other students have completed all the contents and exercises ！ fantastic ！ Thank you for your participation .

In the next blog , I will use a series to explain the contents of each chapter in detail , I hope it can help you to master these knowledge more solidly , I also hope that some of the actual code I share can help you start your own graph data project smoothly , Explore the world of pictures together .

CSDN released The second phase 《 New programmers 》 One of my themes is “ New database Era ： Take you to a new stage of database ”, There are the following descriptions ：

Devour software in open source 、 Today, with the massive amount of Internet data , The database has entered a new stage ： New databases are on the rise 、 Open source database has become a trend 、 Digital transformation is imperative 、 A capital boom gushed out .

Database panorama contains relational database 、 Document oriented database 、 Key value database 、 Graph database 、 Time series database 、 Column store database 、 Search engine database and other open source and closed source products .

I believe that many students who are engaged in software research and development or are interested in doing their own development , Have some knowledge of database . If this is the first time you know “ database ”, You can start with the definition of the database ：

In form ,“ database ” It refers to a set of relevant data and its organization . Access to this data is usually done by a “ Database management system ”（DBMS） Provide , The system consists of a set of integrated computer software , Allow users （ Users of the database ） Interact with one or more databases , And provide access to the data contained in the database .DBMS Provides a variety of functions , Allow input of 、 Store and retrieve large amounts of information , And provide ways to manage how this information is organized .

Let's start from 3 We will start from three aspects .

The history of database development

Here we briefly review the development of databases , Let the students who have not understood the database have a rough Overview , I also want to use this to bring out the development of graph database .

Before the birth of database software system , Data storage and data management have existed for quite a long time . But at that time, data management was mainly through tables 、 Cards, etc , inefficiency , It takes a lot of people to get involved , It is also very error prone .

In this context , On 20 century 60 years , Database system came into being . Let's take a look at some key times and things that appear .

1962 year ：“ database （Database）” The word "R & D" has become popular in the memoranda of system R & D companies .

1970 year ：IBM Of Edgar Frank Codd Write the “ A relational model of a large shared database ” The paper of , This is called “ Father of relational database “.

1979 year ：Oracle Created the first commercial relational database management system （RDBMS）.

1983 year ：IBM Release DB2.

1984 year ：Teradata Release the first massively parallel processing （MPP） Database platform .

1985 year ： Apply to PC The desktop database system appears , Like Microsoft Access.

1988 year ：IBM First put forward “ Data warehouse ” Word and industry standards .

1995 year ： The first open source relational database management system MySQL Release .

2000 year ：Neo4j Our founding team is RDBMS Encountered a performance problem on , And start building the first Neo4j Prototype .

2003 year ： The first paragraph XML Database Publishing , Marks the entry of NoSQL Stage .

2007 year ： The first commercial map database Neo4j Release , The open source version was released at the same time .

2009 year ： Business big data Hadoop platform Cloudera Product launch . Contemporaneous by MongoDB Cause to SQL The wave of .

2010 year ：Neo4j Figure database publishing 1.0 Official version .

2011 year ： Based on resource description framework （RDF） The figure database management system is beginning to emerge .

2014 year ：Spark Release .

2015 year ：Apache Publish more than 25 Data related items .

2016 year ：Neo4j Figure database publishing 3.0 Official version .

In fact, this list can be very long , If you are interested, you can also search the Internet for more detailed historical stories according to keywords . such as Know this information chart .

Classification of database models

Next, let's look at the different categories of database models .

The database model is used to determine the logical structure of the database . It fundamentally determines how data can be stored 、 Organization and operation . A database management system can provide one or more models . The optimal structure depends on the natural organization of the application data , And application requirements , Including transaction rate （ Speed ）、 reliability 、 Maintainability 、 Scalability and cost . Most database management systems are built around a specific data model .

A database model is more than just a way to build data ： It also defines a set of operations that can be performed on the data . for example , The relational model defines SELECT and JOIN Wait for the operation . Although these operations may not be explicit in a particular query language , But they provide the basis for building query languages .

I've been reading 《 Figure database practice 》 This book , There is a comparison chart in Chapter 1 . It can be seen that from the perspective of the widest application , For different data complexity , The database management system can be divided into the following 5 Species category .

Key value （KV） database ： All data has a unique Key As identifier , And the Key Associated data objects . The model is simple and easy to use , Very easy to get started .

Wide column database ： This database system stores data in columns , Each row has a large number of columns representing different data attributes , Allow different rows of data to have different columns .

Document database ： Similar to key value database , There is also a unique key , The stored data is a document organized according to a certain pattern , It can also contain nested data .

Graph database ： chart （Graph） Is a diagram of the data structure , The graph database stores data as nodes （Node） And relationship （Relationship）, Therefore, it has great advantages in processing related data .

relational database ： The oldest and most widely used database model , Store data in the form of a two-dimensional table , Allow associations between different tables .

Let's take a closer look at why the graph database appears , It's not hard to see that although relational databases allow associations , But the association can only be processed during query calculation , Therefore, in the scenario of dealing with multi-level and uncertain associated data, it seems that it is insufficient . Graph database considers both data and relationship from the beginning of model design , Also stored , Therefore, it is very handy when calculating the relationship .

On the other hand , from DB-Engines From the statistical data of the database system on the market , Relational databases have been steadily occupying most of the market , The graph database is the past 10 The fastest growing database type in the year .

About the usage scenario of graph database , I will share several common graph problem patterns in subsequent articles , That is, under what scenario, the graph database has more advantages , Stay tuned .

Looking forward to the future of database

stay 2021 year CSDN 1024 Programmer's Day On ,Neo4j Founder and CEO Emil Eifrem In the round table forum “ Six international database leaders summit dialogue ” On , Shared Figure data platform leads the development of database in the next decade Point of view .

In the past two years, , The development trend of database industry presents three characteristics .

The first is integration , With the large-scale innovation of new data , The rapidly expanding database market once again presents an integrated situation , Hundreds of database vendors are gradually classified as including document databases 、 Graph database 、 Time series database 、NewSQL Four new and stable database camps, including database . In different subdivisions , Only a limited number of companies are singing loudly on the way to lead , As shown in the figure Neo4j And documentation MongoDB.

second , The shift to the cloud has become a long-term trend of the entire industry .2017 Around, some large cloud providers and independent cloud providers began to emerge . Although progress has slowed down due to data gravity and regulatory factors , Now this trend is like a rainbow again . To a large extent, the transfer of data to cloud services has become a driving factor for the development of database platforms .

Third , Data scientists are in the ascendant . As a whole , Data scientists don't like databases , They prefer data . The graph database is an exception , More and more data scientists are interested in graph databases . Data scientists use graphical algorithms to process data , Then input the data into the machine learning pipeline through the diagram , Thus, it provides relational signals for machine learning model and prediction . thus , The graph database becomes the core part of the machine learning pipeline .

on this occasion 2022 On the occasion of the Spring Festival , I believe that with the increasing demand for related data , The popularity of graph databases will soon be a state of affairs . therefore , Now join me in the world of graph data .

I hope that through the introduction of this article , Let you know about the development of the database 、 The category of database model and the development of graph database have a preliminary impression .

Welcome to continue to pay attention and follow up .

Resource Recommendation

• Neo4j The skill tree ： Everyone can learn the learning path of graph database , Help you from 0 Basic mastery Neo4j.
• Become Neo4j Certified Professional ： Many people have passed Neo4j Examination of certified experts . Are you next ？