Avatar of bahard
Flag for Turkey asked on

IO tuning for AIX running oracle ASM


Changing pbuf values (like pv_min_pbuf) has an IO performance change on ASM disks?

Unix OS

Avatar of undefined
Last Comment

8/22/2022 - Mon

No, it hasn't.

pv_min_pbuf is for LVM controlled hdisks (i.e. accessed via logical volumes) and not for raw disks as used by ASM.


ASM uses LVM raw volumes
LVM parameters apply, but since pbufs are auto-tuned by system no care needed there.
More gold is hidden around AIO tunables.

Hello gheist&wmp,

I had known that pv_min_buf does not have a control over ASM disks until I attended a powercare session by IBM and in this training session, the IBM expert said that actually it does. But I have no documentation of it so I am really in doubt about that.

What do you mean by "LVM raw volumes"? When defining ASM disks, we only use disks major and minor number, which means that it is only hdisk-related and there should be no relation with LVM.

And also, this is the first time that I see auto-tune parameters for IBM. If so, could you please explain why I see a regular increase in "blocked_io_count" in "lvmo -a" output. Is there a way to eable auto?

Kind Regards.
Experts Exchange has (a) saved my job multiple times, (b) saved me hours, days, and even weeks of work, and often (c) makes me look like a superhero! This place is MAGIC!
Walt Forbes

I do not know why oe would use PVs, as LVM adds ability to migrate data away from broken hardware while app keeps working....

Log in or sign up to see answer
Become an EE member today7-DAY FREE TRIAL
Members can start a 7-Day Free trial then enjoy unlimited access to the platform
Sign up - Free for 7 days
Learn why we charge membership fees
We get it - no one likes a content blocker. Take one extra minute and find out why we block content.
Not exactly the question you had in mind?
Sign up for an EE membership and get your own personalized solution. With an EE membership, you can ask unlimited troubleshooting, research, or opinion questions.
ask a question

oracle uses AIO to queue requests to storage. if AIO queue is smaller than all pbufs, then obviously they go underutilized... (which is usually the case)