DSXchange

exact difference b/w "peak stage" and "reject link"

through reject link also we can get the rejected data, so why do we go for the "peak stage"(what is the exact functionality of "peak stage").

"hi ray"?

There's more than one person answering questions here. I know it doesn't seem like it at times, but take my word for it, there is.

Peek stage is development/debug stage. It is mainly used for debugging and it can also be used for monitoring the progress of your application.
For more Information refer to Parallel Job Developer's Guide.

Yah, usually helps to read the documentation.

Peek stage is development/debug stage. It is mainly used for debugging and it can also be used for monitoring the progress of your application.
For more Information refer to Parallel Job Developer's Guide.

Please read this topic or, at the very least, its subject line.

There is no such thing as a Peak stage.

There is, as others have noted, a Peek stage. This has the function of capturing a sample of rows, either to log entries or to a file.

A reject link is a link, so that's the main difference.

Many stage types support reject links, the exact function of the reject link is different for different stage types. For example, a reject link on a Sequential File stage with an output traps any rows read from the file that can not be parsed using the record schema associated with that stage's import operator. A reject link on a Lookup stage captures any stream input row that failed the lookup. A reject link on a Transformer stage captures any input row that caused an exception somewhere in the processing of that row. And so on.

There is no single "exact difference", in the same way that there's no exact difference between chalk and cheese.

ray.wurlod wrote:There is no single "exact difference", in the same way that there's no exact difference between chalk and cheese.

I've had some cheeses that... well, never mind.