topic Pseudo swap in Operating System - HP-UX

Pseudo swap

Bhushi — Tue, 10 Jul 2007 10:12:46 GMT

Hi gurus,

Can u plz explain me what is pseudo swap in HP-Ux.

Regds
Bhushi

Re: Pseudo swap

Pete Randall — Tue, 10 Jul 2007 10:19:48 GMT

Steven gives a good explanation here:

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=128060

Pete

Re: Pseudo swap

Oviwan — Tue, 10 Jul 2007 10:20:13 GMT

Hey

check this doc:
http://docs.hp.com/en/5965-4641/ch01s09.html

Regards

p.s. don't forget to assign points! also in your other threads..

Re: Pseudo swap

A. Clay Stephenson — Tue, 10 Jul 2007 10:22:44 GMT

It is not actually swap at all. It is a technique to tell the kernel to allow 75% of physical memory to count as process reservation space.

Consider a box with 16GiB of RAM and 4GiB of device or filesystem swap with pseudoswap disabled. The maximum amount of space for processes would be 4GiB regardless of the amount of free memory. Now if we turn on pseudoswap (swapmem_on=1), 0.75 x 16GiB = 12GiB is added to the reservation space and we can run 16GiB's of processes. Pseudoswap is intended for those systems which have much more memory than swapspace (you bought all that memory so that you would not swap!) and serves no purpose in systems where swap >= RAM (it should be disabled in those cases).

Important concept for UNIX: The amount of PHYSICAL memory does not limit the number and size of processes which can be run; the amount of VIRTUAL memory does. The amount of available physical memory does dramitically alter the speed at which those processes can run.

Re: Pseudo swap

Dave Hutton — Tue, 10 Jul 2007 10:25:14 GMT

From the help within the kernel configuration for swapmem_on:

In previous versions of HP-UX, system configuration required
sufficient physical swap space for the maximum possible number of
processes on the system. This is because HP-UX reserves swap space
for a process when it is created, to ensure that a running process
never needs to be killed due to insufficient swap.

This was difficult, however, for systems needing gigabytes of swap
space with gigabytes of physical memory, and those with workloads
where the entire load would always be in core. This tunable was
created to allow system swap space to be less than core memory. To
accomplish this, a portion of physical memory is set aside as
'pseudo-swap' space. While actual swap space is still available,
processes still reserve all the swap they will need at fork or execute
time from the physical device or file system swap. Once this swap is
completely used, new processes do not reserve swap, and each page
which would have been swapped to the physical device or file system is
instead locked in memory and counted as part of the pseudo-swap space.