Hi Folks,
1.Can anyone please explain me how many types of hash files are there?
2. Can anyone please explain me how to achieve parallelism in server jobs?
Have a nice day
Raj
Hash file types
Moderators: chulett, rschirm, roy
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
What have you already learned from the manuals and from searching? And when's the interview.
There are, in DataStage, two kinds of hashed file. These are called dynamic and static, based on whether or not their internal "table space" management is governed automatically by the volume of data requiring to be stored. I used the word "kind" deliberately there, because "type" is also used to indicate the hashing algorithm associated with a static hashed file, leading some to deduce that there are eighteen different types of hashed file.
Pipeline parallelism is achieved in server jobs through inter-process row buffering, whether or not this is used in conjunction with stage types such as IPC. Partition parallelism is achieved in server jobs through design (multiple paths within a job performing the same tasks) and/or through multi-instance jobs running multiple instances. Whichever you use, the parallelism does not scale other than by manual intervention; one of the main advantages of parallel jobs is that they can scale automatically simply by using a different configuration file.
Oh, you don't have a premium membership yet?![Crying or Very sad :cry:](./images/smilies/icon_cry.gif)
There are, in DataStage, two kinds of hashed file. These are called dynamic and static, based on whether or not their internal "table space" management is governed automatically by the volume of data requiring to be stored. I used the word "kind" deliberately there, because "type" is also used to indicate the hashing algorithm associated with a static hashed file, leading some to deduce that there are eighteen different types of hashed file.
Pipeline parallelism is achieved in server jobs through inter-process row buffering, whether or not this is used in conjunction with stage types such as IPC. Partition parallelism is achieved in server jobs through design (multiple paths within a job performing the same tasks) and/or through multi-instance jobs running multiple instances. Whichever you use, the parallelism does not scale other than by manual intervention; one of the main advantages of parallel jobs is that they can scale automatically simply by using a different configuration file.
Oh, you don't have a premium membership yet?
![Crying or Very sad :cry:](./images/smilies/icon_cry.gif)
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact: