Skip to main content
Announcements
Qlik Connect 2024! Seize endless possibilities! LEARN MORE
cancel
Showing results for 
Search instead for 
Did you mean: 
ravindraa
Creator
Creator

Datamodel

Hi all,

we have 4 database each database have more then 50 tables clients are asking one data model for all database's in this case what is the Pros and cons , if any one knows please give me suggestions .how much ram requiered

present my server Ram size 8 GB.

it is very urgent......................

1 Solution

Accepted Solutions
tresesco
MVP
MVP

I am assuming that your 4 databases are related to four applications and the use wants all of them to be merged as one. This requirement is a common one, since this reduces the required number of Doc CALs. If this is the scenario, one has to understand that Doc CALs might not work for a model that has big(subject to number of records and tables it has) multiple disconnected data islands. Therefore, if you can make an association among the models (i.e. - they are connected), you can logically go ahead with the model. Now comes the data size and performance question. For that, refer to the earlier link.

View solution in original post

3 Replies
tresesco
MVP
MVP

Number of tables would not help to calculate the required RAM. See: Calculate how much RAM needed?

ravindraa
Creator
Creator
Author

Hi tresesco,

  could you share detailed information, you have given help full answer but my requirement is if we design one data model for all application , what is the prerequisites if any one have please give me suggestion.

tresesco
MVP
MVP

I am assuming that your 4 databases are related to four applications and the use wants all of them to be merged as one. This requirement is a common one, since this reduces the required number of Doc CALs. If this is the scenario, one has to understand that Doc CALs might not work for a model that has big(subject to number of records and tables it has) multiple disconnected data islands. Therefore, if you can make an association among the models (i.e. - they are connected), you can logically go ahead with the model. Now comes the data size and performance question. For that, refer to the earlier link.