Which approach is optimized

[SunilChauhan]

hello all,

I have two approaches below

1. My column is increasing each week. i.e for each week date one column

2. second approach i have is  my rows are increasing week by week

each time week data is added and after concatenating its just doubles.

so i would like to know which approach is better for optimization perspective.

Thanks

Sunil

[marcus_sommer]

I'm not sure if I understand it right - every week your data-sources increased with one column? Then you need to transform these data with crosstable to reach a data-stream instead a widely table. And you should consider to use incremental logics.

Incremental Load

- Marcus

[vinay_hg]

Go with row wise data entry. it is the best approch

[senarath]

Hi Guys,

If new column are being added every week, I guess the column names are also different, In which case, you have to load * from table. I wonder how you handle field names of those columns in charts....

So I think, as Vinay..had said, better to re arrange your dbase to row level data.

thanx

[SunilChauhan]

we have historical data and each week we are storing qvd's.it might be  for one year.

it depends on me to model data in such a way that weither row increased or column increase

i am wonder which is best approach for future optimization preferences.

i have also one point that qvd's store columnar data thats why they are optimized.

don't know what is best optimized way.

Sunil