0

Qlik semantic model optionally uses Autonumber for hashes

The hashes created by TimeXtender semantic Qlik model consumes 45 bytes per entry. A Qlik data model using these hashes can consume a lot of memory. In many implementations it is possible to use the Qlik Autonumber feature to drastically reduce the memory footprint (in two sample apps we have reduced by approximately 40%),

It would be nice to have an option which allows us to chose to use the power of Qlik to re-index these hashes in a memory efficient way.

Here is an example of code generated by TimeXtender,

"SuppliersToCompanyName":
LOAD
"SName",
"CompanyName" AS "SuppliersToCompanyName.CompanyName",
"SupplierCode" AS "SuppliersToCompanyName.SupplierCode",
"~SuppliersToCompanyName - TProductLog",
"~PICompanyInfo - SuppliersToCompanyName";
SQL SELECT
"SName",
"CompanyName",
"SupplierCode",
"~SuppliersToCompanyName - TProductLog",
"~PICompanyInfo - SuppliersToCompanyName"
FROM "DirectReliefMDW_Prod"."dbo"."DonorProfiles_DonorProfilesQVDs_SuppliersToCompanyName";

and the same code could be written as:

 

"SuppliersToCompanyName":
LOAD
"SName",
"CompanyName" AS "SuppliersToCompanyName.CompanyName",
"SupplierCode" AS "SuppliersToCompanyName.SupplierCode",
Autonumber("~SuppliersToCompanyName - TProductLog", '~SuppliersToCompanyName - TProductLog') as "~SuppliersToCompanyName - TProductLog"
Autonumber("~PICompanyInfo - SuppliersToCompanyName";'~PICompanyInfo - SuppliersToCompanyName') as "~PICompanyInfo - SuppliersToCompanyName"
SQL SELECT
"SName",
"CompanyName",
"SupplierCode",
"~SuppliersToCompanyName - TProductLog",
"~PICompanyInfo - SuppliersToCompanyName"
FROM "DirectReliefMDW_Prod"."dbo"."DonorProfiles_DonorProfile

 

 

0 comments

Please sign in to leave a comment.