public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Thomas Cataldo <tomc@compaqnet.fr>
To: User & <breno_silva@beta.bandnet.com.br>
Cc: Valdis.Kletnieks@vt.edu, linux-kernel@vger.kernel.org
Subject: Re: Expand VM
Date: 24 Jan 2003 00:20:34 +0100	[thread overview]
Message-ID: <1043364034.25529.18.camel@localhost> (raw)
In-Reply-To: <20030123194014.M374@beta.bandnet.com.br>

On Thu, 2003-01-23 at 20:40, User & wrote:
> Hi Valdis
> 
> Create a new VMA on Linux B for Linux A is easy , but i have a problem , the 
> address of VMA is returned on Linux B , so the VMA created on Linux B can not 
> be used for process of linux A.
> The problem is "how can i return address of VMA created on LINUX B to Linux 
> A , and use this space ?".
> 


What you're looking for is openmosix. It does process migration and so
on..

> Thanks
> Breno
> 
> On Thu, 23 Jan 2003 11:55:38 -0500, Valdis.Kletnieks wrote
> > On Thu, 23 Jan 2003 12:56:27 -0300, User & 
> > <breno_silva@beta.bandnet.com.br>  said:
> > 
> > > I have one idea , and this is about expand virtual memory on linux boxes 
> > > connected in LAN.
> > > Example: Linux A is processing come information , and need more memory , 
> so 
> > > with this source , Linux A could access virtual memory on Linux B in LAN.
> > 
> > We've seen *this* done before (remember diskless Sun3-50's?) - the /dev/swap
> > file would be a large file on an NFS mount from a server.  At the 
> > time, this actually made performance sense, because the old 
> > 'Shoebox' drives the -50 came with were incredibly slow, and you 
> > could actually do an NFS operation to a larger server (a -280 with 
> > Fujitsu SuperEagle disks, for instance) faster than talking to the 
> > local disk.
> > 
> > These days, it's probably easier and cheaper to just buy more RAM 
> > and/or disk for Linux A.
> > 
> > > But i don´t know how translate the virtual address between Linux A and 
> B , to 
> > > have success in acess VM, or how to send all the process for Linux B to 
> be 
> > > processed.
> > 
> > Sending the whole process to Linux B to be processed is called "process
> > migration", and is a difficult problem.  Moving the memory image of the
> > process is usually pretty easy.  What is difficult is moving things like
> > references to open files, file locks, and so on (what if the process 
> > is actively writing to block 739 of /usr/foo/some.file, and the 
> > LinuxB machine doesn't have a /usr/foo, or the permissions on 
> > some.file don't match, or another process has it locked, or... ) 
> > There be nasty dragons in this.
> > 
> > You're probably better off buying more RAM and disk for your A machine.
> > -- 
> > 				Valdis Kletnieks
> > 				Computer Systems Senior Engineer
> > 				Virginia Tech
> 
> 
> 
> ----------------------
> WebMail Bandnet.com.br
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
-- 
Thomas Cataldo <tomc@compaqnet.fr>


  parent reply	other threads:[~2003-01-23 23:16 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-01-23 15:56 Expand VM User &
2003-01-23 16:04 ` Richard B. Johnson
2003-01-23 16:19   ` John Bradford
2003-01-23 16:33     ` Sean Neakums
2003-01-23 16:56       ` John Bradford
2003-01-23 17:58         ` Herman Oosthuysen
2003-01-23 18:46           ` John Bradford
2003-01-23 19:42             ` Herman Oosthuysen
2003-01-23 16:18 ` Patrizio Bruno
2003-01-23 16:55 ` Valdis.Kletnieks
2003-01-23 19:40   ` User &
2003-01-23 20:02     ` Valdis.Kletnieks
2003-01-23 23:20     ` Thomas Cataldo [this message]
2003-01-23 17:09 ` kkonaka
2003-01-24 15:20 ` Bill Davidsen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1043364034.25529.18.camel@localhost \
    --to=tomc@compaqnet.fr \
    --cc=Valdis.Kletnieks@vt.edu \
    --cc=breno_silva@beta.bandnet.com.br \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox