Clients occasionally pause for short periods of time when connected to a large active database
UNIX sync command causes a pause during database Checkpoints

At the time a bi checkpoint is triggered, the bi mechanism must flush all modified database blocks found in the full bi cluster and place them onto the checkpoint queue.

After the checkpoint it will issue a UNIX SYNC to tell the operating system to flush all modified O/S blocks from the UNIX cache. This includes all database pages present in the filesystem cache as  well as other pages that have nothing to do with the database. Any process writing must wait until the filesystem write operations have completed.

With very active databases often with large bi cluster sizes, this task can take a fair amount of time and could become quite noticeable to the clients.  

•  What is a Checkpoint?  

0.  Tune BI I/O Performance and Upgrade for Product BI Performance enhancements: 

•  How to tune BI I/O performance  

In OpenEdge 11.7.6 and 11.2.1, all supported platforms were enhanced to use this new mechanism:

•  Performance improvement when new biclusters are added   

1.  Use the database startup parameter -directio to spread the I/O load.

•  How Does -directio Improve Performance?  

2.  Start more APW's

With or without -directio, enabling more APW's will flush buffers, so there are less buffers to flush at checkpoint time. Assure that a BIW is started and an AIW (if after-imaging is enabled)

•  Why More APWs are Needed When Using -directio?  

3.  Tune the UNIX sync command

Progress 9.1E02 and OpenEdge 10.0B02 Service packs, code was introduced  to use fdatasync rather than sync which ensures the data associated with closed bi clusters is flushed to disk before returning.  There is no longer any need to wait a specific amount of time anymore as previously with the sync call. The -G parameter which used to default to 60 seconds was changed to 0 seconds because of this change.

The UNIX sync command usually runs every 30/60 seconds but may need to run more often to spread the I/O load. More frequent flushes of the file system caches means that the sync() calls issued by the storage manager therefore take less time. Consult with your Operating System vendor on how to do this. One possibility is to have a shell script running in the background, which does something similar to this:
 

while:
do
sleep 5
sync
Done

 

Progress Article:

AIX Performance Issues: -directio and syncd