Abstract

Amazon Neptune is a graph database service that supports two graph models: W3C’s Resource Description Framework (RDF) and Labeled Property Graphs (LPG). Customers choose one or the other model. This choice determines which data modeling features can be used and – perhaps more importantly – which query languages are available. The choice between the two technology stacks is difficult and time consuming. It requires consideration of data modeling aspects, query language features, their adequacy for current and future use cases, as well as developer knowledge. Even in cases where customers evaluate the pros and cons and make a conscious choice that fits their use case, over time we often see requirements from new use cases emerge that could be addressed more easily with a different data model or query language. It is therefore highly desirable that the choice of the query language can be made without consideration of what graph model is chosen and can be easily revised or complemented at a later point. To this end, we advocate and explore the idea of OneGraph (“1G” for short), a single, unified graph data model that embraces both RDF and LPGs. The goal of 1G is to achieve interoperability at both data level, by supporting the co-existence of RDF and LPG in the same database, as well as query level, by enabling queries and updates over the unified data model with a query language of choice. In this paper, we sketch our vision and investigate technical challenges towards a unification of the two graph data models.

中文翻译：

---

OneGraph 愿景：打破图模型锁定的挑战 1

摘要

Amazon Neptune 是一种图形数据库服务，支持两种图形模型：W3C 的资源描述框架 (RDF) 和标记属性图 (LPG)。客户选择一种或另一种型号。这种选择决定了可以使用哪些数据建模功能，以及——也许更重要的是——可以使用哪些查询语言。两种技术栈之间的选择既困难又耗时。它需要考虑数据建模方面、查询语言特性、它们对当前和未来用例的充分性以及开发人员知识。即使在客户评估利弊并做出适合其用例的有意识选择的情况下，随着时间的推移，我们也经常看到新用例的需求出现，可以使用不同的数据模型或查询语言更轻松地解决这些需求。因此，非常希望可以在不考虑选择哪种图形模型的情况下选择查询语言，并且可以在以后轻松修改或补充。为此，我们提倡并探索 OneGraph（简称“1G”）的理念，一种单一的、统一的图数据模型，同时包含 RDF 和 LPG。1G 的目标是通过在同一数据库中支持 RDF 和 LPG 的共存，以及通过使用所选查询语言在统一数据模型上启用查询和更新，实现数据级别和查询级别的互操作性. 在本文中，我们勾画了我们的愿景并研究了统一两个图形数据模型的技术挑战。