Here’s the table of contents:

DTCC | 2021 China map database technology conference link sharing

     In this session DTCC The conference ,Neo4j The company's technical experts will share Neo4j How to meet both functional and performance requirements in terms of features and architecture 、 At the same time, it provides the best scalability . For those interested in graph data technology, please click the link below to learn more . In addition, this article shares some , My own superficial understanding and experience of graph data technology .

One 、 Next generation distributed architecture

     Whether the graph database is suitable for the traditional distributed architecture or Fabric framework ？Fabric thought Is better than traditional distributed architecture Higher dimension Design of distributed architecture , New generation distributed architecture adaptability 、 More scalable , At the same time, the architecture can be compatible with the traditional centralized architecture and distributed architecture 【 Dimensionality reduction in architecture 】. Capacity And performance ？Fabric The architecture can fully inherit the advantages of centralized and distributed storage engines , Connect different storage engines 【 But there are higher requirements for data model consistency , The complexity of operation and maintenance will also be higher 】.

Two 、 Data circulation and data transaction

     data circulation And data transaction scenario , With the help of Fabric The architecture can realize the rapid integration and flow of data . Automatic translation techniques using business models and data models , Achieve model level circulation 、 Build trust and incentives , Can speed up Data standards The real landing .

3、 ... and 、 Business model and data model

     business model and Data model ？ Business modeling related materials ; Data model related information . Business model is a model that can be quickly understood and used by front office business personnel , In traditional data engineering, data engineers need to understand the business before they can accurately model the business translate For the data model , And this process is completely manual . Real data engineering scenarios , Data model and business model are not one-to-one correspondence, but more abstract modeling , Therefore, it is very difficult for business personnel to directly use the underlying data , It is usually necessary to use some business oriented systems to visually obtain 、 Handle 、 Understand the data .

Four 、 Knowledge map data production

     At present, most of the knowledge maps established by enterprises are knowledge maps in vertical fields . combination Fabric The structured data platform introduces blockchain and other technologies to establish a data trust mechanism 、 Incentives , It can accelerate the integration of global data 、 Theoretically, a set of Global knowledge map Platform technology for data construction .

5、 ... and 、 Figure actual combat case of data research center

     In actual production , The problem we use the graph platform to solve is the efficiency of data acquisition in intelligent investment research analysis 、 Data association, etc . Considering the performance 、 cost 、 stability 、 Extensibility 、 Community maturity and other factors , We use based on Neo4j Open source derivative diagram database ONgDB Build a drawing platform , Through secondary development and integration of plug-in tools . It has basically met Deep learning model training Call data and Intelligent search The needs of . Through semi-automatic production tools olab-etl Realize continuous and efficient production of graph data ; adopt Federated query technology Realization MySQL、Oracle、SQL server、Elasticsearch、MongoDB、Neo4j、ONgDB And so on , The integrated data can be used Cypher Standardized interface query , At the same time, it realizes the external storage of engine data Virtual graph translation （ All storage engines go the same way Query Interface standards , The preliminary realization of One Query standard ）.

     Show a sample workbench image （ be based on ONgDB The secondary development workbench and other basic drawing data tools are used as the basic support ）, Figure the data engineer uses the standard Cypher Query Interface can realize graph database 、 Traditional data warehouse 、 Search engine and other storage engine data acquisition , That is, data from other places can be obtained without switching out of the platform （ The return result can be a list 、 Virtual graph and other formats ）; Analytical queries can be called with one click , It can improve the efficiency of data processing and analysis .

6、 ... and 、 Recommended links

Figure special session analysis of database technology |DTCC2021 Conference on database technology of Chinese cartography

Technology trends | Decentralized knowledge mapping collaboration platform construction practice