Archive

Archive for the ‘SSIS’ Category

avoid time-out for Denodo ADO Net Source in SSIS

In SSIS, I was executing sql command in ADO Net Source of one of the Data Flow and was getting Time-out error after approximately 15 minutes. I used below clause after select statement.

context (‘querytimeout’ = ‘9000000’)

Here time is in milliseconds (in our case, ‘9000000’ is approx 2.5 hr).

For more info read “CONTEXT CLAUSE” in Denodo documentation in “Advanced VQL Guide”

Performance Tuning in SSIS for Denodo ADO Net Source

Assumption: You have Denodo driver installed on server. (I had used “DenodoODBC Unicode” version 9.03.04.00, when I created User DSN)

Below steps helped me import two million rows (6 columns with mostly int data) data in less than 4 minutes. Earlier it used to take more than one and half hour.

Here trick is to change User DSN settings. :):):)

  • Click “Start” and type-in c:\Windows\SysWOW64\odbcad32.exe and hit enter in Search Programs and File to open “ODBC Data Source Administrator”
  • In “ODBC Data Source Administrator”, in “User DSN” tab select appropriate Data source from “User Data Sources” list. Click “Configure”.
  • Above step will open “ODBS Driver setup” window, Click “Datasource” button
  • do below changes in “Advanced Options” window
    1. on Page 1
      1. in “Data Type Options”, un-select “Text as LongVarchar”
      2. in Miscellaneous, change “Chunk Size” from 100 to 75000
    2. on Page 3
      1. in “The Use of LIBPQ library”, change selection from “unspecified” to ”yes” , click Apply, Click Ok.

I hope this helps.

How to achieve Parallelism in SSIS

May 20, 2012 Leave a comment

SQL Server Integration Services (SSIS) allows parallel execution in two different ways

1) MaxConcurrentExecutables : It defines how many tasks (executables) can run simultaneously. It defaults to -1 which is translated to the number of processors plus 2. Please note that if your box has hyperthreading turned on, it is the logical processor rather than the physically present processor that is counted. Read more…