shankar


{ City } pune
< Country > india
* Profession * etl developer
User No # 113786
Total Questions Posted # 0
Total Answers Posted # 2

Total Answers Posted for My Questions # 0
Total Views for My Questions # 0

Users Marked my Answers as Correct # 2
Users Marked my Answers as Wrong # 0
Questions / { shankar }
Questions Answers Category Views Company eMail




Answers / { shankar }

Question { 5374 }

when load type is selected as bulk or normal in session level ?let me know the internal process and give me an example?


Answer

A:when you run a session with Normal load IS commits data into the target table as well as commits rowid into a log table.
For Example: you have 10000 records in your source,while loading data if your session fails at any point of time then there is a scope of recovery.After you fetch the issue when you rerun the job you need to ensure that your session is running with Normal load and instead of loading from the scratch,IS connect to the Repository service then Repo service connect to the Repo Database and check what is the last rowid commited in the log table and start loading from the max rowid+1 on wards.

checkpoint at session level: Resume From Last save heckpoint
BulkLoad: when you run a session with Bulk load the IS loads the data into target table and it won't commit any rowid into the log table. so performance will be good because only one operation is happing here but if session is failed at any point of time then there is no scope of recovery.
Note: when you work with Bulk load you need to ensure that there is no indexes on target table.if it has indexes your session get fails.
Q: I want to use bulk load and i have indexes on target table can't i access?
A: Yes you can achieve this by ensuring the option
Pre-SQL as :Drop Index index_Name;
Post-SQL as:create Index index_Name on table_Name(col_Name);

Is This Answer Correct ?    1 Yes 0 No

Question { 3855 }

what is the difference between repository & Intergration service


Answer

what is the difference between repository & Integration service?

Repository: consist of original data base tables which are also called as system tables and metadata objects.

Ex:Actual source data (customers data,employee data,sales data)

Repository service: Repository service is responsible for inserts,updates,deletes and retrieve metadata from Repository. so it will work like mediator between the Repository and Integration Service(client components).

Ex: if you want to extract the data from Repository(actual data) at run time through Integration service you'll get an error because IS doesn't understand the System tables directly so you will need Repository service to handle inserts,updates,deletes and retrieve the data from Repository.

Integration Service: IS is responsible for executing the ETL Objects and performs Extraction,Transformation and Loading at Run time(while executing the session).IS needs Original database objects using objects we can move the data from source to target as per the requirements.

Ex: if you have a business requirements like calculations and conversions all these done by the IS.

Is This Answer Correct ?    1 Yes 0 No