How to create the interval match and connect it to the master calender

[#1]

Hi all,

There are so many date range is there in my application I don't have any idea how can i use this in my document. Major problem for me is to connect the master calender with the date range(from and to date). There are so many date range field is there in many of the tables of my document. So I confused on if i select the date or range of date how it will fall in this interval. Also all the from and to date in the tables are different.

So as to created the month,year and other dates field I have generated the master calender to date selection. But don't have any idea on how to connect this master calender with the date range and avoiding the synthetic key which may arise here.

Have attached the same document which I am working on. Hope some one can work on it to give me start.

[#2]

Hi Darius,

In my case when I reload the same document sent by you. I am getting the memory error. I have 4GB RAM memory in my system and still getting the memory error. I guess in your system the memory is 2 GB and still you did not get the memory error.

It is true giving away 2 days of your time is not easy I can understand that and I am totally agreeing with you. But can you at least suggest me with one small example with my data on how to remaster the data structure so that I can do the same for the rest. Once this is done I guess it will be easy for us to solve the problem.

As you want to know about the objective of this application. The main objective of this application is that it will be used by the company people to analyze their sales force. I don't think it is required to explain you about the sales force as you are much experienced. So as to explain briefly about this it contains the entire sales force data. My technical team lead is writing the queries to pull the required data from database which is required for analysis. This is I just explained as you wanted to know the over all objective of this application.

Hope you can give some hint for me to start with so that we can over come from this problem.

Waiting for your reply!

[d_pranskus]

Hi Rikab

If you compare the structure of your initial application and the one I sent you recently you'll find that there are several tables joined into one. This allow to simplify the overall structure of the application.

After loading dwr tab,, circular associations occur. You could avoid this by joining several tables (use join command, see QV manual for syntax). If it this does not help, then you should use the same technique I used to create link table.

The last approach could be renaming some fields and thus breaking some circular associations. Use formulas in charts to mimic the presence of these broken links.

Cheers, Darius

[#4]

Hi Darius,

Yes that true! After you had joined few tables the structure was indeed simplified. I will surely try to join the tables in dwr tab to make it more simplified. But I have a doubt here! If join table like this will I be able to do the incremental load? As I wanted to do incremental load for all these tables which you have joined. If incremental load can be done then I think I wont be having any issues in joining all these tables to one. Please let me know your view on this.

• I may not be able to apply link table technique as I am not sure how it works.
• Renaming can be done but as per my technical lead all those fields which are joined and creating the loop problem are required to be connected there is no other way for this as per him.
• Also as suggested formula to mimic the presence of those broken links may be too tough to handle. As we may need to apply those set analysis frequently in almost all the charts.

Please let me know you view so that I can proceed further!

[#5]

Hi Darius,

Still I am in the same stage. Just working on how to simplify the dwr table to avoid the circular reference/loosely coupled table.

By the way I have some doubt. I just wanted to know why there is a memory error though I have 4 GB ram in my machine and where as in your it is only 2 GB(I guess) and still there is no memory related error.

Please let me know what can be done for this so that I can do the same.

[#6]

Hi Darius,

Sorry for the disturbance!

I know you are very busy these days. But please check out the possible cause of the issue(memory related error) and let me know.

[#7]

Dear Rikab ,

Just going through the data model what you given & what Darius come up with .

I can notice few things here which i like to share with you:

1) This looks to be a production Schema , becoz information is in scattered manner . I think here you need to convert it into probable Star Or Snowflake using business information , to get best perofmace.

2) One more approach is to create intermediate docs , which will generate QVDs for your final application doc. By doing so you can also get a level of aggregation which will make your job very easy in final doc.

3) Linking the tables are nothing but joining on common fields & making sure "uncommon fileds" are renamed .

Summary : clubbing relevant tables which unables you to derive correct business information is the need here .

Welcome for any clarifications.

Regards,

Bhushan N

[#8]

1) This looks to be a production Schema , becoz information is in scattered manner . I think here you need to convert it into probable Star Or Snowflake using business information , to get best perofmace.

Problem here is I don't the data model of this project which I am working on. I don't have any clue i,e which table is which and which column is related to which. So it is very difficult for me to make this is one in to star schema or snowflake and I don't have any previous experience on the same.

2) One more approach is to create intermediate docs , which will generate QVDs for your final application doc. By doing so you can also get a level of aggregation which will make your job very easy in final doc.

Can you please guide me how to do the same. I don't have much knowledge about the idea which you have mentioned. Are you talking about incremental load kind of thing here.

3) Linking the tables are nothing but joining on common fields & making sure "uncommon fileds" are renamed .

May I know why uncommon field should be renamed?

Summary : clubbing relevant tables which unables you to derive correct business information is the need here .

Is it possible for you to do to. Really I am not able to understand where to start with.

[#9]

Hi Darius,

I need your help very badly here! Please give some time of yours for me and try to help me out. I am not able to overcome from this problem. There is no progress after that. I am still there where we were before.

Hope to receive your reply soon!

[#10]

Darius Pranskus wrote:
You are getting no loops because I have commented dwr page in script before sending it to you.
To help you solve the problem I need to understand the requirements of the application and it looks that the resulting data structure should be some how remastered. But unfortunately I have no time for that. I could help you in some smaller issues, but not that big.
QV works best when you have one fact table and several dimension tables. In your case there are several fact tables, related to different dimension tables. So it causes circular references. <div></div>

Could you please tell me which are all the fact tables in my application. Also let me know what is meant by fact table for my better understanding. I was thinking that all those tables which has from and to date interval are called as fact tables. Correct me if I am wrong!

[#11]

please tell me step by step how to create market calender