Showing results for 
Search instead for 
Did you mean: 

Swapping VS Paging

Regular Advisor

Swapping VS Paging

Hi Friends,
Again with some queries,so please ...
- What is the actual diff between swapping & paging?
on hp unix 11.0, when I issued the command vmstat -s I got the following output,
3195 swap ins
3195 swap outs
71 pages swapped in
72 pages swapped out
37928404 total address trans. faults taken
16786895 page ins
91413 page outs
498372 pages paged in
141174 pages paged ou

Is 3195 above in os blocks/pages?
and one more thing, How to find out the os block size?

Thanks & Regards
Let's share the great thing "THE KNOWLEDGE"
S.K. Chan
Honored Contributor

Re: Swapping VS Paging

Check below for swapping vs paging ..,,0x917187dc4d7dd5118ff00090279cd0f9,00.html

To find out the blocksize ..
1) For JFS file system run ..(example)
# fstyp -v /dev/vg01/lvol2
Look for "f_bsize"
2) For HFS file system ..
# tunefs -v /dev/vg01/lvol2 | grep bsize

Peter Kloetgen
Esteemed Contributor

Re: Swapping VS Paging

Hi Syed,

the difference between swapping and paging is the following:

swapping -->> outsourcing whole processes into
swapping areas, which are not used for a longer period when RAM- space is needed.

paging -->> taking single pages of actually 4KB size which are not used for a longer period and put them into the swapping areas.

All UNIX- processes reserve swap-space, which doesn't mean that swapping/paging is actually done.

Allways stay on the bright side of life!

I'm learning here as well as helping
Vijeesh CTK
Trusted Contributor

Re: Swapping VS Paging

Mark van Hassel
Respected Contributor

Re: Swapping VS Paging

Just a remark on the JFS blocksize:

(Strangely) the f_frsize from fstyp -v gives the (logical) blocksize for a JFS file system. You van test this with creating an FS with the -b option. By the way the default blocksize is 1KB for file systems up to 8GB, 2 KB for FS 16 GB etc etc with a maximum of 8KB.

The surest sign that life exists elsewhere in the universe is that none of it has tried to contact us