HPE GreenLake Administration
- Community Home
- >
- Servers and Operating Systems
- >
- Operating Systems
- >
- Operating System - HP-UX
- >
- 11.23 Does pseudo swap compete with the dynamic bu...
Operating System - HP-UX
1830048
Members
21741
Online
109998
Solutions
Forums
Categories
Company
Local Language
back
Forums
Discussions
Forums
- Data Protection and Retention
- Entry Storage Systems
- Legacy
- Midrange and Enterprise Storage
- Storage Networking
- HPE Nimble Storage
Discussions
Forums
Discussions
Discussions
Discussions
Forums
Discussions
back
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
- BladeSystem Infrastructure and Application Solutions
- Appliance Servers
- Alpha Servers
- BackOffice Products
- Internet Products
- HPE 9000 and HPE e3000 Servers
- Networking
- Netservers
- Secure OS Software for Linux
- Server Management (Insight Manager 7)
- Windows Server 2003
- Operating System - Tru64 Unix
- ProLiant Deployment and Provisioning
- Linux-Based Community / Regional
- Microsoft System Center Integration
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Blogs
Information
Community
Resources
Community Language
Language
Forums
Blogs
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
11-19-2007 08:44 AM
11-19-2007 08:44 AM
11.23 Does pseudo swap compete with the dynamic buffer cache?
HP-UX 11.23 Itanium
There is a lot of disk I/O activity on our system, and there is a relationship between the size of the buffer cache and overall performance. Larger buffer cache = better overall performance. The buffer cache is dynamic on this system, i.e. dbc_min_pct is not equal to dbc_max_pct. Also swapmem_on is 1.
The machine has 12 GB of RAM. There is also 12GB of device swap configured at the moment.
Outside of unusual circumstances, there is not any device (disk) paging activity. However, there is enough memory pressure that the buffer cache does not get close to dbc_max_pct.
We have plenty of free disk space on this machine. I can create sufficient device swap, and set swapmem_on to 0. Will this have any effect on the size of the dynamic buffer cache? The goal is to move the memory reservation to the device swap, and use the RAM for buffer cache instead of pseudo swap.
I need opinions from those who know more about HP-UX internals than I do.
What do you think? Is this idea worth a test?
There is a lot of disk I/O activity on our system, and there is a relationship between the size of the buffer cache and overall performance. Larger buffer cache = better overall performance. The buffer cache is dynamic on this system, i.e. dbc_min_pct is not equal to dbc_max_pct. Also swapmem_on is 1.
The machine has 12 GB of RAM. There is also 12GB of device swap configured at the moment.
Outside of unusual circumstances, there is not any device (disk) paging activity. However, there is enough memory pressure that the buffer cache does not get close to dbc_max_pct.
We have plenty of free disk space on this machine. I can create sufficient device swap, and set swapmem_on to 0. Will this have any effect on the size of the dynamic buffer cache? The goal is to move the memory reservation to the device swap, and use the RAM for buffer cache instead of pseudo swap.
I need opinions from those who know more about HP-UX internals than I do.
What do you think? Is this idea worth a test?
3 REPLIES 3
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
11-19-2007 09:16 AM
11-19-2007 09:16 AM
Re: 11.23 Does pseudo swap compete with the dynamic buffer cache?
In your case, since device swap >= physical memory, pseudoswap serves no purpose. It is really intended for boxes which have less swap than physical memory --- and it really isn't swap at all. It's just kernel bookkeeping. In your case, it will certainly do no harm to turn off pseudoswap thogh I don't expect any performance gains either.
The memory reservation doesn't really move either; again, it's kernel arithmatic. The space is still reserved. Bear in mind that you could have a huge buffer cache and still have very high physical i/o rates. It depends upon how much of your data is being accessed multiple times and how your overall data volumes compare to buffer cache size.
The memory reservation doesn't really move either; again, it's kernel arithmatic. The space is still reserved. Bear in mind that you could have a huge buffer cache and still have very high physical i/o rates. It depends upon how much of your data is being accessed multiple times and how your overall data volumes compare to buffer cache size.
If it ain't broke, I can fix that.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
11-19-2007 09:47 AM
11-19-2007 09:47 AM
Re: 11.23 Does pseudo swap compete with the dynamic buffer cache?
Thanks Clay. I understand that a large buffer cache does not necessarily mean low disk activity.
Are you saying that you expect swapmem_on = 0 to have no impact on the size of the dynamic buffer cache?
What are the disadvantages to swapmem_on = 0 for the system I described?
Are you saying that you expect swapmem_on = 0 to have no impact on the size of the dynamic buffer cache?
What are the disadvantages to swapmem_on = 0 for the system I described?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
11-19-2007 01:22 PM
11-19-2007 01:22 PM
Re: 11.23 Does pseudo swap compete with the dynamic buffer cache?
Rodney,
Refer to this link http://h21007.www2.hp.com/portal/download/files/unprot/devresource/Docs/TechPapers/UXPerfCookBook.pdf
for explanation why you need to turn on swapmemon=1
WK
Problem never ends, you must know how to fix it
The opinions expressed above are the personal opinions of the authors, not of Hewlett Packard Enterprise. By using this site, you accept the Terms of Use and Rules of Participation.
Company
Events and news
Customer resources
© Copyright 2025 Hewlett Packard Enterprise Development LP