- Community Home
- >
- Servers and Operating Systems
- >
- Legacy
- >
- Operating System - Tru64 Unix
- >
- Virtual Memory tuning
Operating System - Tru64 Unix
1748268
Members
3659
Online
108760
Solutions
Forums
Categories
Company
Local Language
юдл
back
Forums
Discussions
Forums
- Data Protection and Retention
- Entry Storage Systems
- Legacy
- Midrange and Enterprise Storage
- Storage Networking
- HPE Nimble Storage
Discussions
Discussions
Discussions
Forums
Forums
Discussions
юдл
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
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
тАО08-07-2005 10:17 PM
тАО08-07-2005 10:17 PM
Virtual Memory tuning
I am trying to tune Tru64 system for a 3rd party application that seems to map its input files to virtual memory and access (read) them sequentially.
My questions are:
Does the kernel parameter vm_max_rdpgio_kluster affect also mapped files regions page-in? (In the documentation they only refer anonymous memory.)
What is the algorithm for the number of pages that are swapped-in when a page-fault occurs?
Thanks
Rafi Moor
My questions are:
Does the kernel parameter vm_max_rdpgio_kluster affect also mapped files regions page-in? (In the documentation they only refer anonymous memory.)
What is the algorithm for the number of pages that are swapped-in when a page-fault occurs?
Thanks
Rafi Moor
2 REPLIES 2
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО08-08-2005 12:30 AM
тАО08-08-2005 12:30 AM
Re: Virtual Memory tuning
It happens almost with all the processes,i will suggest you to monitor your VM usage and also monitor the iostat and per process requirement.
What is the swap you are using on your system.
Regards
Manish
What is the swap you are using on your system.
Regards
Manish
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО08-08-2005 07:04 PM
тАО08-08-2005 07:04 PM
Re: Virtual Memory tuning
Thank you Manish,
Naturally I did monitor the system├в s virtual memory. The picture is like that: There is a lot of free memory, a lot of free swap, zero page-out, quite a bit of page-in and page-faults. The processors spend 40-50% in system mode. I assume this is due to heavy page-in from the mapped data files (I am not talking about text regions).
What I want to do is to make the system map more pages for every page-fault. Since the mapped files are accessed sequentially, this might reduce page-faults.
I know mapped files are loaded to the UBC in chunks of ubc_kluster_cnt pages at a time, but there is still short fault for every page that is not in the active/inactive lists.
I am still waiting for answers to my questions.
Thanks
Rafi
Naturally I did monitor the system├в s virtual memory. The picture is like that: There is a lot of free memory, a lot of free swap, zero page-out, quite a bit of page-in and page-faults. The processors spend 40-50% in system mode. I assume this is due to heavy page-in from the mapped data files (I am not talking about text regions).
What I want to do is to make the system map more pages for every page-fault. Since the mapped files are accessed sequentially, this might reduce page-faults.
I know mapped files are loaded to the UBC in chunks of ubc_kluster_cnt pages at a time, but there is still short fault for every page that is not in the active/inactive lists.
I am still waiting for answers to my questions.
Thanks
Rafi
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.
News and Events
Support
© Copyright 2024 Hewlett Packard Enterprise Development LP