Memory Increasing

[Cameron Simpson]

On 01Mar2007 10:28, varun galande <varun.galande at bhartitelesoft.com> wrote:
| I am using Redhat Linux  (AS) kernel version 2.4.27.ELSMP.
| I am facing one issue in the memory segment's for the system,i am using
| nearly about 4 GB of RAM memory but all the segments of it are getting full
| in nearly around two days m not able to figure it out where the memory
| leakage is .M not able to see any process which is consuming the CPU or RAM
| memory.

It's probably mostly block I/O cache. When data is read off disc it is
kept in RAM because reading from disc is very slow. Over time, more and
more will accrue in the RAM, ready for reuse if those files' data are
needed again. Given enough time, most of your RAM will be used for this.

This is not a performance problem - it is an advantage, because it
maximises your chance of having the data _already_ in RAM when it is
next desired.

It is also not a problem for having to "make RAM available" when it is
needed for something else. Data read from a file and stored in a memory
buffer is "clean" data - it is that same as what is on disc, and so does
not need to be put back there. Such a buffer may be _instantly_ discarded
and used for another purpose.

So most of this "used RAM" is "clean", so it is as though it were all
unused, should more memory be needed.

Cheers,
-- 
Cameron Simpson <cs at zip.com.au> DoD#743
http://www.cskk.ezoshosting.com/cs/

The more I know about the WIN32 API the more I dislike it. It is complex and
for the most part poorly designed, inconsistent, and poorly documented.
- David Korn