HPE GreenLake Administration
- Community Home
- >
- Servers and Operating Systems
- >
- Operating Systems
- >
- Operating System - HP-UX
- >
- swapinfo reserve used total
Operating System - HP-UX
1833770
Members
2449
Online
110063
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
Go to solution
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
08-13-2003 11:59 AM
08-13-2003 11:59 AM
I have some questions about swapinfo and memory reservations in HPUX 11.0.
I've read the man page for swapinfo and it does not specify in detail what is being counted in the reserve used total shown.
Can anyone fill me in on what the OS is adding up to arrive at the reserved total shown by swapinfo? What I am getting at is when I go into gpm I can see the RSS and VSS totals for the processes. Is the reserved total shown by swapinfo simply a summation of the RSS and VSS totals for every process currently running?
Also, if I drill down into the Process Memory Regions for a process, I can see the breakdown of the RSS and VSS totals for that process. Those totals include shared memory allocations.
How does the OS count those shared memory allocations when it comes to the swapinfo reservation? Does a shared memory allocation get counted only once because it's shared memory that several processes might be using? Or does each process's shared memory total get added into the overall reservation separately so the OS has each allocation accounted for in case it has to swap the whole process?
I can see how either approach would make sense, but I have not found an answer anywhere.
With gratitude,
Camaron
I've read the man page for swapinfo and it does not specify in detail what is being counted in the reserve used total shown.
Can anyone fill me in on what the OS is adding up to arrive at the reserved total shown by swapinfo? What I am getting at is when I go into gpm I can see the RSS and VSS totals for the processes. Is the reserved total shown by swapinfo simply a summation of the RSS and VSS totals for every process currently running?
Also, if I drill down into the Process Memory Regions for a process, I can see the breakdown of the RSS and VSS totals for that process. Those totals include shared memory allocations.
How does the OS count those shared memory allocations when it comes to the swapinfo reservation? Does a shared memory allocation get counted only once because it's shared memory that several processes might be using? Or does each process's shared memory total get added into the overall reservation separately so the OS has each allocation accounted for in case it has to swap the whole process?
I can see how either approach would make sense, but I have not found an answer anywhere.
With gratitude,
Camaron
Solved! Go to Solution.
3 REPLIES 3
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
08-13-2003 12:06 PM
08-13-2003 12:06 PM
Re: swapinfo reserve used total
The swapinfo reserved total should be a total of all process VSS sizes. RSS is not included seperately in the reserved count, because it is a subset of the VSS size.
As for shared memory, that makes things a little bit harder. Shared memory segments are mapped into a separate part of the address space, and I think they are counted as part of the VSS size that you see, but not the RSS size.
HTH.
As for shared memory, that makes things a little bit harder. Shared memory segments are mapped into a separate part of the address space, and I think they are counted as part of the VSS size that you see, but not the RSS size.
HTH.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
08-13-2003 12:14 PM
08-13-2003 12:14 PM
Re: swapinfo reserve used total
Sorry for somehow getting this into the wrong category. I could swear I was in the system admin forum when I clicked the new message button.
C-
C-
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
08-13-2003 12:41 PM
08-13-2003 12:41 PM
Solution
Hi Camaron,
The reserve field is incremented when a process "reserves" a certain amount of memory in case it later needs to page any part of it out. This may seem a simplistic statement but that is the theory behind it. For instance, if you malloc() an amount of memory the space is not used immediately - it is reserved by the process for future use.
Where you can reference from here is the pstat() interface, which is what swapinfo and Glance use to find the information you are seeing. You will note there are per-process calls and global calls. For each process gpm will show the shared memory addresses irrespective of other processes, it can't really do anything else. For swapinfo's global count it doesn't add up every attach to a segment, that would create a false picture.
Take a look at a previous thread I updated, using a more cli method to illustrate :
http://forums.itrc.hp.com/cm/QuestionAnswer/1,,0xb4da506d69a7d711abdc0090277a778c,00.html
If you have any more questions, please ask.
Cheers,
James.
The reserve field is incremented when a process "reserves" a certain amount of memory in case it later needs to page any part of it out. This may seem a simplistic statement but that is the theory behind it. For instance, if you malloc() an amount of memory the space is not used immediately - it is reserved by the process for future use.
Where you can reference from here is the pstat() interface, which is what swapinfo and Glance use to find the information you are seeing. You will note there are per-process calls and global calls. For each process gpm will show the shared memory addresses irrespective of other processes, it can't really do anything else. For swapinfo's global count it doesn't add up every attach to a segment, that would create a false picture.
Take a look at a previous thread I updated, using a more cli method to illustrate :
http://forums.itrc.hp.com/cm/QuestionAnswer/1,,0xb4da506d69a7d711abdc0090277a778c,00.html
If you have any more questions, please ask.
Cheers,
James.
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