Help Center/ DataArts Studio/ Best Practices/ Automatic Construction and Analysis of Graph Data/ Scenario
Updated on 2024-08-30 GMT+08:00
View PDF
Share

Scenario

Graph Engine Service (GES) is a service for querying and analyzing graph-structure data based on relationships. It is specifically suited for scenarios requiring analysis of rich relationship data, including social relationship analysis, marketing and recommendations, public opinions and social listening, information communication, and anti-fraud.

In DataArts Studio, you can convert raw data tables into standard vertex data sets and edge data sets based on GES data import requirements, and periodically import graph data (vertex data sets, edge data sets, and metadata) to GES using the automatic metadata generation function, perform visualized graphical analysis on the latest data in GES.

Scenario Description

In this case, DataArts Studio is used to convert the user and rating data of a movie website stored in MySQL database to standard vertex and edge data sets, and to synchronize the data to OBS and MRS Hive. Then, the Import GES node automatically generates metadata and imports the graph data to GES.

Figure 1 Business scenarios
Note that the GES graph data formats include vertex data set, edge data set, and metadata. If your raw data is not in these formats, convert it to these formats.
  • Vertex data sets store vertex data.
  • Edge data sets store edge data.
  • Metadata is used to describe the formats of data in vertex and edge data sets.

For details about GES concepts and graph data, see Graph Data Formats.

Constraints

When metadata is automatically generated using the Import GES node, note the following constraints:
  1. You can only select a vertex table and a edge table that use a single label. If you select a vertex table or a edge table that has multiple labels, the generated metadata may be missing.
  2. The metadata XML file is generated after you click Create. If the structure of the vertex table and edge table changes during subsequent job scheduling, the metadata XML file will not be updated automatically. In this case, you need to open the New dialog box and click Create again to generate a new metadata XML file.
  3. In the generated metadata XML file, the value of Cardinality (data composite type) in Property is single and cannot be changed.
  4. You can generate metadata XML files for multiple pairs of vertex tables and edge tables at a time. However, only one table can be selected for the Edge Data Set and Vertex Data Set parameters of the Import GES node. If there are multiple pairs of vertex tables and edge tables, you are advised to create metadata XML files on multiple Import GES nodes. In this way, you can ensure that each piece of metadata corresponds to each pair of vertex tables and edge tables during the import of graph data.