Can anyone explain what the role of these parameters are in a dynamic HF. the DS documentation is not much of a help thanks.
Minimum modulus
Merge load
Split load
Merge load and Large records
All the help appreciated , Many thanks !
Dynamic Hash file options!
Moderators: chulett, rschirm, roy
This looks like a good general explanation:
http://openqm.com/HTMLobj-79/Dynamic_Files.pdf
http://openqm.com/HTMLobj-79/Dynamic_Files.pdf
-
ray.wurlod
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
Not bad as far as it goes, but it doesn't really explain the terms. They are all values that can be set when the hashed file is created or subsequently when the hashed file is tuned.
MINIMUM.MODULUS is the smallest number of groups that the hashed file is permitted to have. Its default value is 1. The hashed file can grow larger than this value, but can not shrink to a smaller size. Used to pre-allocate disk space and to preserve disk space when purging a hashed file.
SPLIT.LOAD is a threshold, expressed as a percentage. Its default value is 80. When the hashed file exceeds this percent full, it "splits", growing a new group.
MERGE.LOAD is also a threshold, expressed as a percentage. Its default value is 50. When the hashed file falls below this percent full, records from its highest-numbered group are merged back into a smaller structure and the highest-numbered group is removed.
LARGE.RECORD is yet another threshold, expressed either as a number of bytes or as a percentage. It represents the size of record that will be handled as oversized. An oversized record has some (most) of its data stored in the overflow portion of the hashed file.
GROUP.SIZE is specified as 1 or 2, the only available choices for DataStage dynamic hashed files. This value indicates the size of every group buffer in the hashed file, and is expressed in units of 2048 bytes.
DataStage hashed files' data are stored in two files called DATA.30 and OVER.30 (not ~0 and ~1). Indexes on DataStage hashed files are not stored in the same directory as the data, but in a separate directory called I_hashedfilename, and using a different naming convention from that given in the cited paper.
MINIMUM.MODULUS is the smallest number of groups that the hashed file is permitted to have. Its default value is 1. The hashed file can grow larger than this value, but can not shrink to a smaller size. Used to pre-allocate disk space and to preserve disk space when purging a hashed file.
SPLIT.LOAD is a threshold, expressed as a percentage. Its default value is 80. When the hashed file exceeds this percent full, it "splits", growing a new group.
MERGE.LOAD is also a threshold, expressed as a percentage. Its default value is 50. When the hashed file falls below this percent full, records from its highest-numbered group are merged back into a smaller structure and the highest-numbered group is removed.
LARGE.RECORD is yet another threshold, expressed either as a number of bytes or as a percentage. It represents the size of record that will be handled as oversized. An oversized record has some (most) of its data stored in the overflow portion of the hashed file.
GROUP.SIZE is specified as 1 or 2, the only available choices for DataStage dynamic hashed files. This value indicates the size of every group buffer in the hashed file, and is expressed in units of 2048 bytes.
DataStage hashed files' data are stored in two files called DATA.30 and OVER.30 (not ~0 and ~1). Indexes on DataStage hashed files are not stored in the same directory as the data, but in a separate directory called I_hashedfilename, and using a different naming convention from that given in the cited paper.
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.
hi robi,trobinson wrote:This looks like a good general explanation:
http://openqm.com/HTMLobj-79/Dynamic_Files.pdf
i couldn get the file from the path that you have mentioned. could you help me out to get the same?
Thanks in advance
veera..
Wasn't all that hard to find where it had moved to:
http://www.openqm.org/downloads/dynamic_files.pdf
Realize that this is a technical paper from a completely different product.
http://www.openqm.org/downloads/dynamic_files.pdf
Realize that this is a technical paper from a completely different product.
-craig
"You can never have too many knives" -- Logan Nine Fingers
"You can never have too many knives" -- Logan Nine Fingers