Ubuntu kernel memory management

[Joel Bryan Juliano]

You can use sysctl to tweak the running kernel,
and /etc/sysctl.conf to apply these settings permanently.

Example,

sudo /sbin/sysctl -w vm.min_free_kbytes=16384

other options are available by invoking -a.

here are the memory options with their default value.

vm.swap_token_timeout = 300
vm.legacy_va_layout = 0
vm.vfs_cache_pressure = 100
vm.block_dump = 0
vm.laptop_mode = 0
vm.max_map_count = 65536
vm.min_free_kbytes = 2039
vm.lowmem_reserve_ratio = 256   256     32
vm.swappiness = 60
vm.nr_pdflush_threads = 2
vm.dirty_expire_centisecs = 3000
vm.dirty_writeback_centisecs = 500
vm.dirty_ratio = 40
vm.dirty_background_ratio = 10
vm.page-cluster = 3
vm.overcommit_ratio = 50
vm.overcommit_memory = 0


On 5/18/06, Riaan Bredenkamp <riaan at shuttleworthfoundation.org> wrote:
>
> Currently running Ubuntu on a ltsp server, X86 architecture, 4 GiB RAM.
> Free reports the following:
>
> admin at tuxlab:~$ free
>             total       used       free     shared    buffers     cached
> Mem:       3108492    1339980    1768512          0     222352     808256
> -/+ buffers/cache:     309372    2799120
> Swap:      2650684          0    2650684
>
> using:
>
> admin at tuxlab:~$ uname -a
> Linux tuxlab 2.6.10-6-686-smp #1 SMP Wed May 3 17:18:12 UTC 2006 i686
> GNU/Linux
>
> So, it seems like the Kernel is splitting the virtual memory into the
> traditional 3G user-space 1G kernel-space, as is described at:
> http://kerneltrap.org/node/2450
>
> * Is this the most optimal way to physical memory available to the system?
> * If not, what would be and how would I go about implementing it?
>
>
> Thanks
> Riaan
>
>
> --
> ubuntu-users mailing list
> ubuntu-users at lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-users
>



-- 
Linux is a gun that you point to Microsoft.
- Joel Juliano
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ubuntu.com/archives/ubuntu-users/attachments/20060518/39f5b213/attachment.html>