RE: OOM band-aid (just a small idea)...

Jesper Juhl (jesper.juhl@dif.dk)
Fri, 9 Oct 1998 11:31:32 +0200


This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------ =_NextPart_001_01BDF367.9B39723A
Content-Type: text/plain

Hello people,

I will not pretend that I know much about the way that Linux handles
memory, but I have a thought that might be of some use (I hope) to
somebody working on the virtual memory.

When we are about to run out of swap, why don't we then start
compressing all data in the swap partition(s), and compress any new data
before it's written to swap (and ofcourse decompress it when it's read
from swap). You would ofcourse need to keep track of the compression
ratio, so that as soon as we can stop using compressed swap we do so.
I know this does not solve the problem. But at least it might be able to
let the system live a little longer, or even make it survive the load.

I hope this will be of some use to somebody.

Best regards
Jesper Juhl

------ =_NextPart_001_01BDF367.9B39723A
Content-Type: text/html
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
RE: OOM band-aid (just a small idea)...

Hello people,

I will not pretend that I know much about the way = that Linux handles memory, but I have a thought that might be of some = use (I hope) to somebody working on the virtual memory.

When we are about to run out of swap, why don't we = then start compressing all data in the swap partition(s), and compress = any new data before it's written to swap (and ofcourse decompress it = when it's read from swap). You would ofcourse need to keep track of the = compression ratio, so that as soon as we can stop using compressed swap = we do so.

I know this does not solve the problem. But at least = it might be able to let the system live a little longer, or even make = it survive the load.

I hope this will be of some use to somebody.

Best regards
Jesper Juhl

------ =_NextPart_001_01BDF367.9B39723A--

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/