Max No of Columns in Open Hub Destination

[Former Member]

Hi Experts,

I wan to know the limitation of number of rows & Columns for OHD output when I use the below as my source.

Be1.Bex as source – Max rows & Columns we can export.

2. MP/Cube as source –Limitation of no of columns we can have and row.

Thanks,

Balaji K S

[ccc_ccc · Accepted solution]

Hi Balaji,

Bex as source/ MP/Cube as source – Max rows & Columns we can export.

  if your target is = Direct DSO (No limit)

                             = Excel file (Columns = 16384, Rows = 1048576) version MS office 2010

                             = Infoobject (no limit)

      And also you can aggregate data by using aggregate transformation (to reduce no of records)

Thank you,

Nanda

[Former Member · Accepted solution]

Hi Nanda,

Thanks for your reply,

In Bex, At max we can add 50 infobjects in the row, in the same way how many info objects we can add it in OHD from Multiprovider/Cube/DSO.

Thanks,

Balaji

[Former Member]

Balaji - Not sure where you are going with this question.  Here is what I had noticed over the years with OHD's.

- They can be as big as your DSO or IOBJ in terms of # of columns.

Now if you create a Flat File with a seperator ( Tab space or a delimitor) Yes you can put atleast 100+ IOBJ up there.

The bigger your flat file based OHD - It may execute slowly since no parallel processing is available for flat files and also based on network traffic.

Multi providers(MPROV) and OHD - This is possible, but there could be a possibility of duplicate data. Also if the incoming loads for the MPROV are not loaded the outbounds OHD will be delayed.

Finally - If your OHD's have say large volume Eg. 200-600 Million rows and about 100+ IOBJ when doing a full load or start up this has to be planned carefully during go live to avoid any downtime to the business users using this data.

[Former Member]

Yes Thanks Krishna.

[ccc_ccc]

Hi Balaji,

if your question cleared, close this thread by selecting "useful" or "Correct answer".

Thank you,

Nanda