如何在不使用 ER 表示法的情况下呈现数据模型
我手头有一个有趣的问题。我在 Erwin 中有一个大型模型 - 我的逻辑数据模型几乎处于物理模型的级别。因此，为了向业务分析师和程序员展示模型，我需要创建一个看起来像业务对象模型的逻辑数据模型。我发现大多数人很难理解 ER 表示法所表达的关系。即使经过大量解释，人们仍然对 ER 表示法感到不舒服，因为它颠倒了我们表达实体的正常方式。
例如，如果我有一个员工和多个地址，在类模型中，我们将把员工作为一个类，将地址作为集合实体。而在数据模型中，地址将具有从员工到地址的基数为 1 到多的 FK 关系。
我想找到一种以图表结构呈现数据模型的方法，如果您愿意的话，可以轻松地以自上而下的方式理解数据模型。
关于我可以使用哪些工具/想法有什么建议吗？我在 Erwin 和 SQL Server 中提供了完整的模型。我可以访问 Erwin、Vision、Microsoft Visual Studio 和 Enterprise Architect
I have an interesting problem on hand. I have a large model in Erwin - My logical data model is almost at the level of physical model. Hence, to present the model to the business analysts and programmers, I needed to create a Logical Data Model that looks like a Business Object Model. I find that most people find it difficult to understand the relationships expressed in ER notation. Even after lot of explanation, people don't get feel comfortable with the ER notation which reverses the normal way that we express an entity.
For example, if I have an employee and multiple addresses, in a class model, we will have employee as a class with addresses as a collection entity. Whereas in the data model, address will have a FK relationship with a cardinality of 1 to many from employee to address.
I want to find a way to present the data model in a diagram structure which makes it easy to understand in a top-down fashion, if you will.
Any suggestions here on what tools/ideas that I can use? I have the full model available in Erwin and in SQL Server. I have access to Erwin, Vision, Microsoft Visual Studio and Enterprise Architect
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
对象角色建模非常适合独立于实现的概念模型。 ORM 是比 ERD 更好的模型共享和交流媒介。能够首先创建概念模型，然后使用 规范。
Object Role Modelling is ideal for conceptual models independent of implementation. ORM is a better medium for sharing and communicating models than ERDs. It's also very useful to be able to create the conceptual model first and then generate the database from it using a tool like NORMA.