Graph database has a complete advantage in dealing with correlation, especially in the Internet era where social networks have been greatly developed.
If we want to know who LIKES who (LIKES can be one-way or two-way), we also want to know who is LEADER_OF everyone. The search takes a huge amount of time to compare one by one, but the subgraph query of the graph database can provide a more accurate response.
Graph databases have the following advantages:
1. The user can think object-oriented, and every query used by the user has explicit semantics, so that the user does not have to worry about how to write the code.
2. Users can update and query the graph database in real time, and flexibly respond to massive relationship changes, such as adding and deleting relationships, entities, etc.
3. The graph database is conducive to the visualization of real-time big data mining results.
The following recommend some good graph data analysis engines:
- GraphScope：
GraphScope is a one-stop graph computing platform developed and open source by alibaba Damo Intelligent computing Laboratory. Relying on ali’s massive data and rich scenarios, as well as damo institute’s high-level research, GraphScope is committed to providing one-stop and efficient solutions to the above challenges of graph calculation in actual production. GraphScope provides a Python client that can be easily connected to upstream and downstream workflows. It has the characteristics of one-stop, easy development, and extreme performance. It features efficient cross-engine memory management, Gremlin distributed compilation optimization for the first time in the industry, automatic parallelization of algorithms and auto-quantization for dynamic graph updates, providing extreme performance for enterprise-level scenarios. In both internal and external applications at Alibaba, GraphScope has proven to deliver significant new business value in several key Internet areas such as risk control, ecommerce recommendation, advertising, cyber security, knowledge mapping, etc.
2. GDB：
GDB (Graph Database) is a standard Graph data product on Ali clound, which is currently in public test. According to the official website, the official commercial time is December 31, 2019. GDB is a real time and reliable online database that supports attribute graph model and is used to process highly connected data query and storage. It supports TinkerPop Gremlin query language. Highly optimized self-mapping computing layer and storage layer, cloud disk multiple copies to ensure ultra-high reliability of data, ACID transaction support; Supports high availability (HA) instances and rapid node failover to ensure service continuity. Provides rich O&M functions, such as backup and restoration, automatic upgrade, alarm monitoring, and failover, greatly reducing o&M costs.
3. Neo4j：
Neo4j is an embedded, disc-based Java persistence engine with full transactional features, but it stores structured data on the network rather than in tables. Neo4j provides large-scale extensibility to handle diagrams of billions of nodes/relationships/attributes on one machine and can be extended to multiple machines running in parallel. Compared to relational databases, graph databases are good at handling large amounts of complex, interconnected, low-structured data that changes rapidly and requires frequent queries — in relational databases, these queries lead to large numbers of table joins and therefore performance problems. Neo4j addresses the problem of performance degradation that occurs when querying a traditional RDBMS with a large number of connections. By modeling the data around the graph, Neo4j traverses the nodes and edges at the same speed, with no relation to the amount of data that makes up the graph. In addition, Neo4j provides very fast graph algorithms, recommendation systems, and OLAP-style analysis, all of which are not available in current RDBMS systems.