Graph database and application scenarios

Graph database and application scenarios

Welcome everyone to Tencent Cloud + community to get more Tencent's massive technical practice dry goods~

This article was published in the cloud + community column by Tencent Cloud database TencentDB

In recent years, enterprise cloud migration has become a trend and trend. The Henan Provincial Government has also issued the "Henan Province "Enterprise Cloud" Action Plan (2018-2020)". How to use the cloud well and how to use the underlying cloud database is also Become a new subject.

Mr. Shao Zongwen, Deputy Director of Tencent Cloud Database Products, will use the industry experience and customer service cases accumulated by Tencent Cloud Database for many years, combined with the comparison of cloud database and self-built database, and share the development trend of database industry analysis with you, and help companies improve industry competition with the help of graph database Strengthen the rapid development of enterprises!

1. Market analysis

The rapidly growing graph database

A picture is worth a thousand words

Compared with traditional information storage and organization models, graph databases can clearly reveal complex models, especially in the intricate social, logistics, and financial risk control industries.

There are many graph databases, including Operational graph database, RDF graph database, multi-mode graph data, analysis and large graph database. Graph databases are getting more and more attention, and most of them are constantly being updated.

Popular king neo4j has collected tens of millions of downloads

The download volume of neo4j is 10M+: **From the perspective of the number of downloads, the prospects are still great, **7M is based on the docker version of neo4j

50K+ of experienced engineers using neo4j: **Through years of training and gradual promotion,** there are currently more than 50,000

The number of participants in the graph database is 50K+: **Graph Database related technical conferences, **GraphConnect conference has thousands of participants.

2. application analysis

After learning so many advanced technologies, how to combine them with business? The graph database covers many industry scenarios and is especially suitable for social and financial risk control fields.

Application case: Financial risk control scenarios involve many dimensions, and traditional databases cannot solve them well

Bank risk pressure mainly comes from: compliance supervision risk, black property fraud risk, and internal employee risk. Taking advantage of the Tupu network, various bank risks can be prevented through four risk management methods:

Staff management

Employees and relatives, employees and external companies

Customer Management

Financial status, credit investigation, industry, liquidity of funds, major events

Relationship management

Family relationship, equity relationship, group relationship, business relationship, supply chain, industry relationship

Business management

Business process, business compliance, business funds, business progress, business data

Tencent's self-developed graph database storage and computing solution, Star Atlas, provides a distributed engine that integrates storage, query, and calculation of super-large graph databases, helping various industries to solve their pain points.

Based on the star map, relevant information and potential laws can be quickly queried, improving data control and avoiding potential risks.

3. Comparison of advantages and disadvantages

At present, the mainstream database used in the industry is still a relational database. What are the advantages and disadvantages of graph databases compared to relational databases?

The graph database has a complete advantage in dealing with the relationship, especially in the Internet era when our social network has been greatly developed. For example, we want to know who LIKES (likes) who (likes can be one-way or two-way), and we also want to know who is whose FRIEND_OF (friend) and who is everyone's LEADER_OF (leader). In addition to its obvious advantages in relational queries, graph databases have the following advantages:

a) Users can think object-oriented, and every query used by users has explicit semantics;

b) Users can update and query the graph database in real time;

c) The graph database can flexibly respond to massive relationship changes, such as adding and deleting relationships, entities, etc.;

d) The graph database is conducive to the visualization of real-time big data mining results.

Although the graph database has made up for many of the shortcomings of relational databases, it still has some shortcomings, such as :

a) Not suitable for recording large amounts of event-based data (such as log entries);

b) Binary data storage.

c) Projects that require high concurrent performance.

d) At present, there are many query languages for related graphs, and they have not been well unified.

e) Some books and documents related to the graph database are insufficient, and the related ecology is still being improved.

Compared with relational databases, there is basically no big difference between graph data and relational databases in the face of conventional queries. In the comparison of complex queries, the graph database is more concise and concise to quickly meet user needs.

4. Industry Outlook

From the survey, many industries have implemented the graph database as a plan. In recent years, among the world's top 100 companies, the proportion of using graph databases has gradually increased. In the financial sector, graph databases are used to achieve anti-fraud and other requirements, and graph databases are widely used in software, logistics, new retail, airlines, telecommunications, hospitals, and biopharmaceuticals.

The graph database is good at handling large, complex, interconnected, and changeable network data, and its efficiency is much higher than that of traditional relational databases a hundred, thousand, or even ten thousand times. The graph database is particularly suitable for a wide range of fields such as social networks, real-time recommendations, banking transaction loops, and financial credit reporting systems. Globally renowned companies such as LinkedIn, Wal-Mart, CISCO, HP, and eBay are all using the graph database Neo4j, and Chinese companies are gradually starting to use the graph database to build their own applications.

If you have not touched or used the graph database, I believe you will definitely touch or use it in the near future! Tencent Cloud is willing to work with people from all walks of life to make a better database world hand in hand!

Related Reading

Learn about the necessary "good medicine" for team battles!

No longer have to worry that the black teammates in the Internet cafe will not be able to hear it!

3 lines of code, add voice interaction capabilities to QQ light games

[Daily course recommendation] Machine learning combat! Quick start online advertising business and CTR corresponding knowledge

This article has been authorized by the author to publish it by Tencent Cloud + Community. For more original text, please click

Search and follow the public account "Yunjia Community", get technical dry goods the first time, and reply 1024 after paying attention to give you a technical course gift package!

Massive technical practical experience, all in the Yunjia community !