Memory Buffer Architecture:
A fundamental understanding of the memory management architecture in SSIS will help you make intelligent choices about how to assist SSIS to move data more efficiently through the pipeline. Some transformations require data to be copied or moved from one buffer to another while others do not. Because some transformations depend on the output of others, this can create blocks and bottlenecks in the process. Some operations require that a preceding transformation completely finish its work before it can begin to process any rows. This is said to be a blocking transformation. Other transformations must begin their duty, but rows can begin to flow to another transformation before the entire process has completed. In this case, the operation of the first transformation only partially blocks the second. Finally, there can be data flows and transformations in the package that are not at all dependent on each other and do not block other transformations in the data flow.
Buffer Usage:
Behind the scenes, the Data Pipeline engine uses a buffer-oriented architecture to load and manipulate data sets in memory efficiently. The benefit of this in-memory processing is that you do not need to copy and stage data physically at each step of the data integration. Rather, the Data Pipeline engine manipulates data as it is transferred from source to destination.
No comments:
Post a Comment