All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jacob Gorm Hansen <jacobg@diku.dk>
To: Kip Macy <kmacy@fsmware.com>
Cc: Jerome Brown <jerome@concepts.net.nz>, xen-devel@lists.sourceforge.net
Subject: Re: uClibc
Date: Tue, 21 Dec 2004 16:20:13 +0100	[thread overview]
Message-ID: <41C83F2D.5050501@diku.dk> (raw)
In-Reply-To: <20041221143457.P10500@demos.bsdclusters.com>

Kip Macy wrote:

> At the risk of starting the construction of a bikeshed, what kind of
> hardware are you running on? The reason for the increased appeal of
> virtual machines over the past few years is that hw performance and
> memory availability has increased faster than applications' needs.

I am running on 2.4GHz Dell Optiplexes with 512MB of memory and no swap 
configured for dom0. Currently I have 16 of them, but in the long run I 
am hoping to run this on 1000+ nodes with similar config. That means any 
memory I waste I can multiply with an arbitrarily large number to make 
it sound scary.

My approach has (to me) many other nice properties, apart from the 
memory footprint, some of the more important ones being security (much 
reduced trusted computing base) and performance+isolation (not using 
shadow page tables, not relying on an external checkpointing service in 
dom0, being able to delay scheduling badly behaving processes during 
migration, knowing who is in my ARP cache, etc.), as well as overall 
better alignment with Saltzer's end-to-end argument.

I would like to create a platform where anyone can purchase resources, 
without any kind of trust that customer are playing nice, and I think 
that I am on the right path.

Before Xen came along, I created a similar platform (called NomadBIOS), 
based on L4, sporting what I now call hosted or managed migration. The 
system I am building now is my 'second system', for better or worse ;-) 
  I probably spent more time thinking about these issues than most 
people (though that is no guarantee that my conclusions are correct).

 > -Kip

Jacob


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now. 
http://productguide.itmanagersjournal.com/

  reply	other threads:[~2004-12-21 15:20 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-12-21 20:14 uClibc Jerome Brown
2004-12-21 20:36 ` uClibc Jacob Gorm Hansen
2004-12-21 20:52   ` uClibc Jerome Brown
2004-12-21 21:11     ` uClibc Jacob Gorm Hansen
2004-12-21 22:44       ` uClibc Kip Macy
2004-12-21 15:20         ` Jacob Gorm Hansen [this message]
2004-12-21 23:36         ` uClibc Jacob Gorm Hansen
2004-12-22  1:48     ` uClibc Mark Williamson
2015-12-19  9:18 uclibc Markus Volk
2015-12-19  9:34 ` uclibc Khem Raj

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=41C83F2D.5050501@diku.dk \
    --to=jacobg@diku.dk \
    --cc=jerome@concepts.net.nz \
    --cc=kmacy@fsmware.com \
    --cc=xen-devel@lists.sourceforge.net \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.