7 Replies Latest reply: May 13, 2014 7:40 AM by Francisco Rogério Almeida dos Santos

# Set Analysis

Good morning all,

I have a problem in Sept analysis, I would like some help to improve the performance of the same, I have several expressions within my analysis would like an opinion to improve this analysis.

I am currently performing the operations necessary in my analysis directly in the script, and using only 2 fields in the front end, someone has to know a little better this performance?

Before was thus the expression

sum ({<CLASSIF_CLI_FAT_DRE = {'test'}, CONTRATO_NOVO = {'test'}

-

sum ({<CLASSIF_CLI_FAT_DRE = {'test'}, {CODIGO_PRODUTO_FAT_DRE = 'test', 'test'}

Now the expression is so because the calculation is being done in the script

sum ({<line {'LINHA10'}, EXPRESSION =} = {'EXPRESSAO1'}> VALUE)

-

sum ({<line {'LINHA10'}, EXPRESSION =} = {'EXPRESSAO2'}> VALUE)

• ###### Re: Set Analysis

Set analysis can't be used in the script. You will have to use if statements instead. I have no idea what you're actually trying to sum. The statements you posted make no sense to me.

• ###### Re: Set Analysis

I'll try to improve my question .
Before was thus the sum expression
( { < CLASSIF_CLI_FAT_DRE = { 'test' } , CONTRATO_NOVO = { 'test' }
- Sum ( { < CLASSIF_CLI_FAT_DRE = { 'test' } , { CODIGO_PRODUTO_FAT_DRE = 'test' , 'test' }

Script is loaded in the next row.

/ / Linha10
/ / EXPRESSA01
/ / ( { < CLASSIF_CLI_FAT_DRE = { 'test' } , CONTRATO_NOVO = { 'test' } +

Concatenate

' LINHA10 ' AS LINHA,
' EXPRESSAO1' AS ESPRESSAO , CHAVE_OCC_LANC ,
CODIGO_EMPRESA_ THE COMPANY
CENTRO_CUSTO_LANCAMENTO THE CENTER ,
TIPO_CCUSTO_CONTABIL AS TYPE ,
ANO_CONTABIL AS ANO_DRE ,
MES_CONTABIL AS MES_DRE ,
VALOR_LANCAMENTO AS VALUE ,
CLASSIFICATION AS CLASSIFICACAO_DRE
RESIDENT LANCAMENTO_CONTABIL
WHERE
CLASSIF_CLI_FAT_DRE = 'test' , CONTRATO_NOVO = 'test' ;

/ / Linha10
/ / EXPRESSA02
/ / Sum ( { < CLASSIF_CLI_FAT_DRE = { 'test' } , { CODIGO_PRODUTO_FAT_DRE = 'test' , 'test' }

Concatenate

' LINHA10 ' AS LINHA,
' EXPRESSAO2 ' AS EXPRESSAO , CHAVE_OCC_LANC ,
CODIGO_EMPRESA_ THE COMPANY
CENTRO_CUSTO_LANCAMENTO THE CENTER ,
TIPO_CCUSTO_CONTABIL AS TYPE ,
ANO_CONTABIL AS ANO_DRE ,
MES_CONTABIL AS MES_DRE ,
VALOR_LANCAMENTO AS VALUE ,
CLASSIFICATION AS CLASSIFICACAO_DRE
RESIDENT LANCAMENTO_CONTABIL
WHERE
CLASSIF_CLI_FAT_DRE = 'test' and match ( CODIGO_PRODUTO_FAT_DRE , 'test' , 'test' )

So on the front end I'm using the following .

sum ( { <line {'LINHA10'}, EXPRESSION = } = {'EXPRESSAO1'}> VALUE )
-
sum ( { <line {'LINHA10'}, EXPRESSION = } = {'EXPRESSAO2'}> VALUE )

But there are not only these two expressions on the front end there are over 100.

I hope you understand

• ###### Re: Set Analysis

Rogerio,

i understand that you have application with bigger data volume and you need performance optimisation.

Maybe there is somebody who will be able to tell you how to do it, but from my experience it may be difficult. In my last project i worked about 2-3 weeks on optimisation on data model. I had 800 mln rows in application. Of course it is very important for app performance how your data are oganised. Sometimes adding some 0/1 flag can help, sometimes it is worth t ojoin some tables together. In general - simpleier model and simpleir expressions works faster

regards

Darek

• ###### Re: Set Analysis

Darek,

My application has around 20 million only, but the server my client is not robust, thanks for the tip I'll try to find something more about ways you commented.

• ###### Re: Set Analysis

Rogerio,

maybe it is good idea to attache sample application with your case.

regards

Daerk