We here have been trying out the online (inplace) reorg feature of v8.2 UDB. As part of this we have been doing application testing and system monitoring.
One of the things noted, is the following:
1. We note that a lot of locking is going on
2. We note that there is a lot of transaction log activity
3. We note no significant changes to the filespace (SMS tbsp) where the tables are stored.
All of the above implies that most of the work is being done by using the transaction log.
I am looking for a more technical explanation of how this thing works. I have found no good explanation through all of my searches.
Does anyone have a good internal explanation of reorg? good links?
![[pacman]](/data/assets/smilies/pacman.gif "[pacman] [pacman]")
Paul
One of the things noted, is the following:
1. We note that a lot of locking is going on
2. We note that there is a lot of transaction log activity
3. We note no significant changes to the filespace (SMS tbsp) where the tables are stored.
All of the above implies that most of the work is being done by using the transaction log.
I am looking for a more technical explanation of how this thing works. I have found no good explanation through all of my searches.
Does anyone have a good internal explanation of reorg? good links?
Paul