ssas/power bi 中有 2 个暗淡日期有问题吗？
我有一个带有事实表和两个维度（其中之一是日期维度）的基本模型。
现在，一个带有日期的新列已添加到事实表中......因此我创建了第二个“暗淡日期”并连接到它： 
我有下一个疑问：
如果我使用 2 个暗日期，我的 .pbix 或多维数据集会出现任何问题吗？
我是否应该将这个新的“暗淡日期”也标记为“标记为日期表”？我可以有 2 个表标记为日期表吗？ 
这个新的“暗淡日期”仅用作 pbix 中的过滤器，我不打算在其上使用任何时间智能...
I have a basic model with a fact table, and 2 dimensions (one of them is Date dimension).
Now, a new column with a date has been added to the fact table… Therefore I have created a second ‘Dim Date’ and connected to it:
I have the next doubts:
Can I have any problem in my .pbix or cube if I use 2 dim dates?
Shall I mark this new ‘dim date’ also as ‘Mark as date table’? can I have 2 tables marked as date table?
This new 'Dim Date' shall be used only as a filter in the pbix, I dont plan on using any time intelligence on it...
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）
这取决于：
power bi 运行的分析服务表格引擎支持表之间的多个连接。我通常建议将其与 USERELATIONSHIP() 函数一起使用，然后您的测量将为报告提供上下文。
但是，我发现在某些情况下，在许多度量中使用 USERELATIONSHIP() 可能会给模型带来不必要的复杂性。您最终可能会得到太多的度量，并且当您使用在同一视觉中使用两种不同关系的两个度量时，可能会变得混乱。
简而言之：复制维度没有任何本质上的错误，但为了数据存储优化和模型清洁，我确信在复制维度之前，具有事实和维度之间多种关系的 USERELATIONSHIP() 将不起作用。
It depends:
The analysis services tabular engine that power bi runs on supports multiple connections between tables. I would generally recommend using this with the USERELATIONSHIP() function and then your measures will give context to the report.
However, I have found there are situations where using USERELATIONSHIP() in many measures can introduce unnecessary complexity in your model. You can end up with far too many measures and it can get confusing when you use two measures that are using two different relationships in the same visual.
In short: There is not anything inherently wrong with duplicating a dimension but for data storage optimization and model cleanliness I would be sure USERELATIONSHIP() with multiple relationships between fact and dimension will NOT work before duplicating the dimension.