[Vserver] Interesting times ...

From: Herbert Poetzl <herbert_at_13thfloor.at>
Date: Wed 21 Dec 2005 - 15:24:57 GMT
Message-ID: <20051221152457.GC28006@MAIL.13thfloor.at>

Hi Community!

for those who do not know yet, I recently was invited
from the PlanetLab folks to pay a visit to Princeton,
which I did, and besides the fact that it was a lot of
fun, we managed to have a bunch of very productive
discussions ...

I'm going to mention some of the ideas (or solutions)
which will make it into Linux-VServer (or already made
it into the devel branch) sooner or later.
(not all of them are the direct result of my visit,
but I guess it doesn't really matter)

 - I/O schedulers come in 4 flavours (noop, deadline,
   antcipatory and complete fair queuing (CFQ)), we
   decided to focus on the CFQ (which now also supports
   priorities) and made that per context (in devel)

 - the CPU scheduler will get an overhaul and will
   become a hybrid Fair Scheduling partially Work
   Preserving multi CPU scheduler :)

   this will allow you to do things like:

   - hand out CPU 'guarantees' (per unit)
   - define a fair share independantly
   - restrict a context to a certain maximum

 - we will continue to develop ngnet and try to make
   it work side by side with the current legacy net
   (well, an updated version of that, at least)

   this will give you:

   - a virtual switch/router like setup, similar to
     UML or real machines, on the host (which then
     basically becomes the router)
   - completely isolated loopback and userspace device
     support for tunneling and similar
   - ipv6 support inside a guest
   - guest per interface accounting (also on the host)
 
 - pid and other types of 'isolation' will be extended
   to do full virtualization without increasing the
   overhead (in cooperation with folks from columbia)

   this is a prerequisite to context migration and
   snapshoting in a cooperation with folks from the
   Columbia university

 - Private Namespaces become hierarchical, i.e. they
   start to propagate certain events, like mount or
   unmount (if desired) down the hierarchy ...

   this is a mainline 'feature', but it will become
   very interesting for Linux-VServer I guess ...

 - We will look into creating a high speed kernel
   userspace interface to query/poll/report status
   information for graphing all kinds of stats easily

 - Dynamic Context support will be removed from the
   kernel, and moved into userspace.

 - Persistent Contexts (even without processes inside)
   are now possible

 - Context Setup can be done from inside (SETUP state)
   or from the outside

all folks interested in helping with one or the
other sub-project please contact me and give me
some details about your plans ...

all folks interested in testing one or the other
feature, or willing to provide some infrastructure
for the folks going to test/develop that, please
send an email and let us know ...

everybody interested in the features and/or similar
features please follow up to this thread so that
we can prioritize them properly ...

ah, and I almost forgot, if you want to sponsor any
of those developments (which would probably set
some priority and allow us to get more work done),
please let us know ...

thanks a lot,
Herbert

_______________________________________________
Vserver mailing list
Vserver@list.linux-vserver.org
http://list.linux-vserver.org/mailman/listinfo/vserver
Received on Wed Dec 21 15:25:22 2005

[Next/Previous Months] [Main vserver Project Homepage] [Howto Subscribe/Unsubscribe] [Paul Sladen's vserver stuff]
Generated on Wed 21 Dec 2005 - 15:25:28 GMT by hypermail 2.1.8