SSIS transformation tasks can use memory or buffer in different ways. The way transformations use memory can impact the performance of your SSIS package dramatically. Transformations buffer usage can be categorised as below.
Non blocking transformations
Partial blocking transformations
Fully Blocking Transformations
If you clearly observe Sort is a fully blocking transformation, so its better to sort your data using the SQL command in OLE DB Source instead of using Sort transformation. Merge tranform requires Sort but not Union All, so use Union All wherever possible.
Hope this post helps in analysing the performance bottle necks in your SSIS package and also while developing new SSIS package for performance tuning.
Reference: DP (http://dptechnicalblog.blogspot.com/ )
- Non blocking transformations
- Partially blocking transformations
- Full Blocking transformations
Non blocking transformations
- Audit
- Cache Transform
- Character Map
- Conditional Split
- Copy Column
- Data Conversion
- Devired Column
- Export Column
- Import Column
- Lookup
- Multicast
- OLE DB Command
- Percentage Sampling
- Script Component
- Slowly Changing Dimesion
Partial blocking transformations
- Data Mining
- Merge
- Merge Join
- Pivot
- Unpivot
- Term Lookup
Fully Blocking Transformations
- Aggregate
- Fuzzy grouping
- Fuzzy lookup
- Row Sampling
- Sort
- Term Extraction
If you clearly observe Sort is a fully blocking transformation, so its better to sort your data using the SQL command in OLE DB Source instead of using Sort transformation. Merge tranform requires Sort but not Union All, so use Union All wherever possible.
Hope this post helps in analysing the performance bottle necks in your SSIS package and also while developing new SSIS package for performance tuning.
Reference: DP (http://dptechnicalblog.blogspot.com/ )
No comments:
Post a Comment