Hello all Ab Initio developers,I have recently graduated and I have just joined the Data Warehousing profession, I got some basic training in Ab Intio for few months now and have a few questions, can someone please help me. 1) How does parallelism work in Ab Initio? When and how do you decided when to use the different types of parallelism?1. Component -- multiple components running on SEPRATE data at the SAME time.2. Pipe-line -- Multiple components running simultaneously on SAME data at the SAME time.3. Data -- data divided into sections and operates on all sections simultaneously.2) How do you select the number of parallelism for Data parallelism? Example: 2-way, 3-way 8-way? 3) What is the number of parallelism Ab Intio can support?4) When and what is the method used to design Ab Initio graphs?5) What is a staging area? Is a staging area required in ETL all the time?6) How does OLAP and ETL work together?Thanks you all, April (Anmol)

Questions by AprilBlue   answers by AprilBlue

Showing Answers 1 - 3 of 3 Answers

ankithak

  • Oct 13th, 2007
 


 1)     In abinitio component,pipeline parallesim supports default.so the user cant change it. so for data parallesim supports only when ur  graph has muitifiles with partition or departion componets  only.

2) IT depends upon the degree of parrallesim.
3)Default 2 ie componet,pipeline
4) Method depends upon ur requirement .No need to follow any particular method.
5)Before loading to the data into dwh we need staging area.The main advantage is
if u want to get the recent data no need to connect dwh we can get from staging.

6) before olap we have to do etl after that to make reports we need olap.
if u have any doubts regarding send me
i will make it clear

  Was this answer useful?  Yes

Give your answer:

If you think the above answer is not correct, Please select a reason and add your answer below.

 

Related Answered Questions

 

Related Open Questions