Microsoft SQL Server 2005 Integration Services

"QUICKER THAN A GREASED WEASEL."

EUAN GARDEN

The Data Flow Task is performant by design such that out of the box, without any tuning or optimization, the default settings generally deliver great performance. For less complex data flows, that is data flows with five or six components with small data loads of a gigabyte or less, the default settings are typically sufficient for most applications and do not require any special tuning. However, large and complex data flow loads bring their own challenges and there are often more trade-offs and decisions that need to be made that the Data Flow Task cannot make for the user or over which it has no control.

This chapter starts by describing important Data Flow Task concepts you should understand for the tuning discussion and then proceeds to describe specific ways you can get even more performance using simple settings. These are called turnkey settings because there isn't a lot of measurement or process involved when deciding to use them and the benefits from the settings can usually be seen immediately. The section "An Approach to Data Flow Optimization" discusses a general approach you can use for identifying and resolving performance issues and squeezing every last ounce of performance from the Data Flow Task.