OE - 11.7.2 running on Red Hat OS
Is this Enterprise RDBMS or Workgroup RDBMS? If it is Workgroup, you shouldn't be using a large BI cluster size. If you care about tuning for performance, you should be using Enterprise.
Will check pointing decrease by increasing AIBUF and BIBUF?
No. A checkpoint happens every time a BI cluster is filled with BI blocks containing change notes. The writing of this data is cached, in BI buffers. Having more BI buffers can help if you see Empty BI Buffer Waits (promon R&D, 2, 5). The frequency of checkpoints is a function of your transaction volume (rate of writes to the BI file) and your BI cluster size. Increasing the BI cluster size will increase time between checkpoints. What are your current BI block size and cluster size? What is your current time between checkpoints during heavy update activity? Why do you think it needs to decrease?
Do I have to do both of them (Understood from articles that their values should match)?
The -aibufs setting should be the same as -bibufs. There was guidance once upon a time to set them differently, e.g. to set -aibufs to 1.5 * -bibufs.
Also, is there a concept of too much memory allocated to aibuf/bibuf?
Yes, there are theoretical and practical limits on lots of parameters, including -bibufs and -aibufs. But a reasonable setting would be nowhere near any such limit. Assuming a BI block size of 16 KB, a value of -bibufs 150 works out to about 2.3 MB of memory. Compared with the big consumers of shared memory like -B, this is a drop in the bucket.
The current values sit at 150 and I am thinking of increasing them to 512 (One of my DB's already have this values).
A value of 150 should be plenty for most workloads. What is your reasoning for considering an increase to -bibufs/-aibufs?
Are you running background processes (BIW, AIW, WDOG, APWs)?