Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi All
Can any one guide me on the following issues, I have had a Qvw file with size 500 MB. I am taking it in to the other qvw file as resident ,
and I am transforming the fact table here,
like
Fact:
LOAD
ACCOUNTID,
AutoNumber(DEPTID&'|'&DOCNUM&'|'&DOCTYPE&'|'&DCDTNUM&'|'&DATE&'|'&ReportID&'|'&FIRMTYPE&'|'&DFGH) AS KEY1,
//ACCOUNTID&'|'&DOCNUM&'|'&DOCTYPE&'|'&DCDTNUM&'|'&DATE&'|'&ReportID&'|'&FIRMTYPE&'|'&DFGH AS Account_Key,
AutoNumber(DOCNUM&'|'&DOCTYPE&'|'&DCDTNUM&'|'&DATE&'|'&ReportID&'|'&FIRMTYPE&'|'&DFGH) AS Key2,
num(trim(DEPTID)) AS DEPTID,
num(trim(GLMCU)) AS GLMCU,
num(trim([BRANCH NUMBER])) AS [BRANCH NUMBER],
num('20'&(FISCALYEAR+1),'##0.0') AS FISCALYEAR,
([SALES]/100) AS [SALES],
Period AS Period,
AddYears('1/1/1900',num(left(DATE,len(DATE)-3)))+num(right(DATE,3))-1 AS [FiscalDate] ,
AutoNumber(num('20'&(FISCALYEAR+1),'##0.0')&Period) AS Key,
Date(AddYears('1/1/1900',num(left(DATE,len(DATE)-3)))+num(right(DATE,3))-1,'MM-DD-YYYY') AS JulianDate,
Date(AddYears('1/1/1900',num(left(DATE,len(DATE)-3)))+num(right(DATE,3))-1,'MM-DD-YYYY') AS DATE,
num(Date(AddYears('1/1/1900',num(left(DATE,len(DATE)-3)))+num(right(DATE,3))-1,'MM-DD-YYYY')) AS DateNum,
DOCTYPE AS [DOC NUMBER],
DESCRIPTION,
[ABAP DESCRIPTION],
FIRMTYPE AS [Account Type],
FIRMTYPE,
BRANCH,
DOCTYPE,
DOCNUM,
num(DCDTNUM) AS DCDTNUM ,
DATE,
ReportID,
DFGH
Resident TEMPFact;
and I am calculating the other aggregation table with
Sum(SALES) AS AGGR_SALES with KEY1,FISCALYEAR,PERIOD etc.
WHEN I run this application I am facing problem with reload time. App taking more time 3 hours how can I improve the reload time here.
Regards
John
20 billion records might be simply too big for 8GB. Take a good look at the ram usage of your machine during reload.
there is something I don't understand
500 mb .qvw
a lot of fields in fact table
20 billion of records
8 gb ram
the numbers of fields and records seems too big for a 500 mb qvw
Sorry Wrong Typo it 20 Million
Sorry Wrong Typo its 20 Million