On 26 Nov 2008, at 10:06, Fritz Zaucker wrote:

> On Tue, 25 Nov 2008, Burak Arslan wrote:
>
>> first, we're not straying too much off-topic, are we? if so please  
>> tell.
>
> Another (border line) comment:
>
> Adding additional components to a system (SW or HW) will in general  
> decrease
> the reliability, as it will add additional bugs and ways for failure.
>
Absolutly right.

> As today's systems are a lot more reliable than some years ago (when  
> WAS
> your last failure of a (reasonable) server disk/CPU/power supply/etc)?
> It might be better to invest in a setup you can easily get back in  
> operation
> if it fails (which usually also means better documentation,  
> automization,
> etc), than on adding additional layers of complexity.
>
I also agree but at that point have to enter some specific  
consideration : it depends on your business constraint.
Say it take t to get a new server from strach up and running, it  
business is fine to wait t, that's fine, if they don't want to wait  
more than t, then they have to put more budget and it ends by  
duplicating servers, just to make sure it will be less than t to wait.  
Of course, it impact the budget.

I think my dev responsability as not to block as far as I can that  
sort of deployment decision.
So, it should be OK to deploy all on one server (and I agree with you  
and Burak this is simpler and have a lot of benefit) or to go up to 3  
tiers physical architecture with servers duplicated and load balancer.
=> That's why I likle the idea of beeing able to have stateless server.
Even on all in one server, this stateless server software architecture  
could be a big bonus : less software component, less memory footprint  
and in case of core dump, just relaunch the process : no context in  
session.

> There is an interesting editorial by Rick Farrow in the last issue  
> of ;login
> (Usenix Journal) talking about the security impact of server  
> virtualization.
> I think his comments apply to the current discussion as well.
>
> Cheers,
> Fritz
> [CUT]
I hope no one is bothered by that discussion I found interesting :-)

JBB.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
qooxdoo-devel mailing list
qooxdoo-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/qooxdoo-devel

Reply via email to