What is new arrive dimension?

[myshah789]

Some more questions please....

Q1. Why do we use determinant? Please explain with examples?

Q2. What is new arrive dimensions? Please explain with examples?

Q3. Suppose we have a table which is holding a large number of records. There are 4 columns which are very frequently used in the Cognos reports, so can we create indexing on those 4 columns?

[MFGF]

Are you sitting some kind of exam? The questions you are asking (here and in your previous post) are from a wide range of topics, and I find it difficult to understand how they could all stem from one report. Some concern cubes, some concern DMR, some concern admin and some concern reporting. If I didn't know better, I would think you were trying to get answers to some kind of certification test.  Can you explain what you are trying to achieve and how these questions fit in?

MF.

[myshah789]

Hi MFGF,
Some of these questions I'm facing from past 2 months and still looking for correct answers.
I'm working on cubes, report studio reports etc. and ofcourse some admin part as well.

I'm a new member of this Forum, first I thought to ask one by one, then I thought let me ask all the questions simaltaneously so I may get all my answers in one go. So I asked them.
I'm really sorry, If I have troubled to my techies. But still hoping that I'll definately get all my answers from here.

Thanks & Regards,
Shah

[blom0344]

But you are aware that the 3 questions in this topic have almost zero relevance to either cubes, report studio and administration?  Asking for examples simply smells of exams.
We'd like to help you with technical issues , though  ;)

[MFGF]

Hi MFGF,
Some of these questions I'm facing from past 2 months and still looking for correct answers.
I'm working on cubes, report studio reports etc. and ofcourse some admin part as well.

I'm a new member of this Forum, first I thought to ask one by one, then I thought let me ask all the questions simaltaneously so I may get all my answers in one go. So I asked them.
I'm really sorry, If I have troubled to my techies. But still hoping that I'll definately get all my answers from here.

Thanks & Regards,
Shah

Ok - that's fine. :)

Some more questions please....

Q1. Why do we use determinant? Please explain with examples?

Q2. What is new arrive dimensions? Please explain with examples?

Q3. Suppose we have a table which is holding a large number of records. There are 4 columns which are very frequently used in the Cognos reports, so can we create indexing on those 4 columns?

Q1. Determinants are used by the query engine to understand how to aggregate multi-grain facts linking to a single conformed dimension.  For example, you might have sales revenue in one fact, which joins to the time dimension at the date level, and monthly sales targets in another fact, which joins to the same time dimension at the month level. If you created a report containing (say) Month, Revenue and Target, the query engine would need to aggregate the measures differently - revenue would need to be aggregated, but target not aggregated in this instance.  Determinants on the time dimension (in this case) allow the query engine to understand these differences and to perform appropriate aggregation.

Q2. I haven't come across "New Arrive Dimensions" per se. I'm wondering if you are referring to Late Arriving Dimensions? If so, these are changes to a dimension table in a data warehouse which are being processed out of sequence (ie not changes to the current dimensional row). Dimension tables often preserve history using Type 2 SCD behaviour, where when (for example) a product gets renamed, a new dimensional row is added for the product, marked as the current row and allocated an Effective Date, and the existing row for the product is marked as old and given an End Date. To preserve uniqueness, surrogate keys are often used in the table, and these are in turn used in fact tables as the basis for the Dimension > Fact relationships. A Late Arriving Dimension gives all sorts of headaches in this situation, because it effectively means the current history trail in the dimension is incorrect - there are missing changes. To fix things, the missing rows need to be inserted in their appropriate positions in the history trail, the Effective/End dates of the rows currently in the table need to be modified to suit, then all the fact tables need to be rebuilt from scratch to regenerate the correct surrogates to link them to the relevant rows in the dimension.

Q3. Yes. This is a database task. Often DBAs will handle these kinds of changes for you. They will also know whether adding indexes is appropriate, and if so what kind of index suits best (B-tree, bitmap index etc).

Cheers!

MF.

[myshah789]

MF, thank you very much.