Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi
we are planning a search dashboard with the data was huge we did some front aggregations like (count(distinct) in the dashboard but for optimization purpose my manger told me to all the aggregations ,if i did back end aggregations my data was getting doubled
Example : Count(DISTINCT if(cdate>= vStartdate and cdate<=vEnddate and stage_no=1, cookie)) the value was 63130
with aggregationed changes in the backend i used this formula
sum(DISTINCT if(cdate>= vStartdate and cdate<=vEnddate and stage_no=1, Search) the value was 124212
it was doubled and it was wrong , i tried so many changes i commented the stages and cookie columns but the result was same
i'm sharing the code which i used
SEARCH:
LOAD
autonumber(cookie & cdate) as cookie,
date(cdate) as cdate,
MonthName(cdate) as c_Month,
channel,
stage,
stage_no,
Booking_no,
test_ids
FROM
(qvd)
where match(stage_no,'1','5','7','8','9','10')
and cookie <>'Unknown' and match(channel,'SITE','MOBILE');
NoConcatenate
SEARCH1:
LOAD
cookie,
cdate,
channel,
stage,
stage_no,
Booking_no,
test_ids
count(DISTINCT if( stage='1', cookie)) as STAGE1,
Count( DISTINCT if( stage=8, cookie)) as STAGE2,
Count( DISTINCT if( stage=9, cookie)) as STAGE3,
Count( DISTINCT if(stage=10, cookie)) as STAGE4,
Count( DISTINCT if( stage=7, cookie)) as STAGE5,
Resident SEARCH
where match(channel,'SITE') AND len(test_ids)>0
group by
apache_cookie,
cookie,
cdate,
c_Week,
c_Week,
channel,
domain,
stage,
stage_no,
sector,
Booking_no,
test_ids
drop table SEARCH;
Regards
Pardhu
I don't even feel performance issue while using this on front-end. Can you share your rows in that object
Count(DISTINCT if(cdate>= vStartdate and cdate<=vEnddate and stage_no=1, cookie))
Changed to below and Try from your end then check how index seems in GDB{Governance Dashboard} - {If available}.
Count({<cdate = {">= $(vStartdate) <= $(vEnddate)"}, stage_no={'1'}>} DISTINCT cookie)
The reason i am feeling change is, Aggregation good and works too in script level. But, It effect to the DM
Actually the data was historical and we are thinking if the data was Aggregation in the backend we are assuming it will decrease dashboard size
True, I agree for your way.
But, In script i didn't feel any wrong. So, I will explain you one thing
Recently(8 Months back) i did Aggregation on script only. After that i removed that Pivot / Straight objects and used Table box for Aggregation fields simply and then It showing Out of memory with X Mark.
Hope clear and helps to you
the set analysis was not working showing zero values
It should work, Can you share the snap where it is not working