Design all dimensions as conformed dimensions
3 posters
Page 1 of 1
Design all dimensions as conformed dimensions
Hello,
what are the pros and cons designing all dimensions as conformed dimensions across the data warehouse. (all fact groups have each and every dimension key and when it's not applicable we can set default value like -1)
Thanks in advance
what are the pros and cons designing all dimensions as conformed dimensions across the data warehouse. (all fact groups have each and every dimension key and when it's not applicable we can set default value like -1)
Thanks in advance
ALW123- Posts : 2
Join date : 2015-06-24
Re: Design all dimensions as conformed dimensions
Conformed dimensions are a key part of dimensional modelling - there are no pros or cons as if you are designing a dimensional model you have to use conformed dimensions; if you don't then it's not a dimensional model.
If what you are asking about is the pros and cons of different DW methodologies - Kimball or Inmon - then that's a different question but this is probably not the best forum to get an unbiased opinion on the topic
If what you are asking about is the pros and cons of different DW methodologies - Kimball or Inmon - then that's a different question but this is probably not the best forum to get an unbiased opinion on the topic
nick_white- Posts : 364
Join date : 2014-01-06
Location : London
Re: Design all dimensions as conformed dimensions
Hi Nick,
I asked what if I design a fact table with all the dimension attributes in data warehouse as an agile practice ? what are the pros and cons
I asked what if I design a fact table with all the dimension attributes in data warehouse as an agile practice ? what are the pros and cons
ALW123- Posts : 2
Join date : 2015-06-24
Re: Design all dimensions as conformed dimensions
Off the top of my head, I can't think of any pros to what you are suggesting...
You should keep the Fact record size as small as possible - as you will normally have high volumes of records in Fact tables. Adding Dimension keys that you will never use to a fact makes little sense to me.
I wouldn't call what you are suggesting an "agile practice", it sounds more like just trying to skip over the design stage.
I guess it might be a workable approach if you had a very low number of facts and dimensions in your DW - but then if you've only got a few facts you might as well design them properly.
I would suggest that if you are going to follow the Kimball methodology (and I'm not saying you have to) then you don't try and re-invent the wheel. This methodology has been evolved over the years and has been proved to work by probably 10s of thousands of successful DW implementations. I would suggest that the reason the methodology doesn't suggest putting all your data in a single fact table surrounded by all your dimensions is because it doesn't work
You should keep the Fact record size as small as possible - as you will normally have high volumes of records in Fact tables. Adding Dimension keys that you will never use to a fact makes little sense to me.
I wouldn't call what you are suggesting an "agile practice", it sounds more like just trying to skip over the design stage.
I guess it might be a workable approach if you had a very low number of facts and dimensions in your DW - but then if you've only got a few facts you might as well design them properly.
I would suggest that if you are going to follow the Kimball methodology (and I'm not saying you have to) then you don't try and re-invent the wheel. This methodology has been evolved over the years and has been proved to work by probably 10s of thousands of successful DW implementations. I would suggest that the reason the methodology doesn't suggest putting all your data in a single fact table surrounded by all your dimensions is because it doesn't work
nick_white- Posts : 364
Join date : 2014-01-06
Location : London
Re: Design all dimensions as conformed dimensions
Fact tables should have the dimensions that provide context to the fact. Nothing more, nothing less.
The term 'conformed dimension' doesn't mean every fact has the same dimensions. It means the dimension is well defined and used consistently in multiple fact tables.
The term 'conformed dimension' doesn't mean every fact has the same dimensions. It means the dimension is well defined and used consistently in multiple fact tables.
Similar topics
» Conformed dimension design
» Conformed dimensions
» Conformed Dimensions
» Conformed Dimensions
» About Conformed dimensions
» Conformed dimensions
» Conformed Dimensions
» Conformed Dimensions
» About Conformed dimensions
Page 1 of 1
Permissions in this forum:
You cannot reply to topics in this forum
|
|