[Gllug] Checkpoint/Restart

[Vidar Hokstad]

On 26 Feb 2008, at 12:48, Mick Farmer wrote:
>
> One of our researchers needs to run a program for a long
> time (probably a week or more), so I thought it might be
> useful if we provided a checkpoint/restart facility.

Apart from application level solutions it's worth considering
virtualization solutions for this. Setting up Xen or OpenVZ
is fairly simple, and both of them support both checkpointing
and migration.

I have good experiences with OpenVZ (we host several large
customers in OpenVZ containers, and have done transparent
failover with the help of OpenVZ when one of our boxes ran
into hardware problems recently).

The upside of OpenVZ is that it's light and fast. The downside
compared to Xen is less isolation (in particular it doesn't let you
control real memory vs. swap on a per container level, while
Xen does), and OpenVZ doesn't support running different
kernels in the containers, as it's containers are just sanboxes
running on the host kernel..

Vidar

-- 
Vidar Hokstad <vidar at aardvarkmedia.co.uk>
Technical Director, Aardvark Media <http://www.aardvarkmedia.co.uk/>




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.lug.org.uk/pipermail/gllug/attachments/20080226/37409c41/attachment.html>
-------------- next part --------------
-- 
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug