Re: VMMs and Over-commitment

[Marcus Brinkmann]

At Sat, 11 Aug 2007 18:05:37 +0200,
<address@hidden> wrote:
> Well, to put it bluntly: Systems with improved interfaces failed in the
> market, while VMs are very successsful...
> 
> I don't mean to say it's hopeless and we should resign. What I mean to
> say is that it can't be all attributed to wrong research focus. There
> must be much more tangible reasons, and it's sure worthwile to
> understand these.
> 
> I can think of two very decisive reasons. One is that with VMs, instead
> of switching to something completely new, you only need a specific bit
> of additional infrastructure (the VM itself), and can otherwise stick
> with all the mature and familiar software they have been using before.
> 
> The other reason is that interfaces allowing for more isolation and/or
> independance in a single system are very abstract; it's hard to
> understand how to use them, or why one wants it. VMs on the other hand
> are a concept that is very easy to grasp: They just provide a sealed
> room within the system, with total isolation and independance, allowing
> to run another system (or another instance of the same system) in there.

I don't think there is an intellectual challenge in understanding
concepts.  The pressures seem to come from elsewhere.  The historical
development of the PC architecture becoming a commodity, and thus
ubiquitous, has entrenched Windows as the only operating system on the
Desktop, and Unix or Windows on the servers.  The switching cost to
something else (*anything* else) is high enough to keep everyone in.
Because everybody understands this, you won't even get funding to try
to change that.  There is not even a market where you could fail.

Virtualization is becoming popular because the hardware of a single
machine now outpowers common loads of a single user significantly.
Thus it becomes feasible and cost-effective to switch to a thin client
model in certain work places (call-centers etc) and to hosting
multiple domains on a single internet service (ISPs).  These
applications themselves justify significant commercial interest in
VMs.  Researchers follow the funding.

However, a new potential market opens up with embedded devices.
Embedded devices are becoming much more powerful and thus include more
applications.  With this, there is pressure on the underlying platform
to adjust.  Note that VMs don't help for these devices, but sandboxing
applications does.  Read the recent seL4 paper from ERTOS/NICTA to get
a glimpse into their point of view.

> It's also wrong to attribute all interest in VMs to isolation alone. The
> independance property seems just as important.

VMs are a major opportunity to get hosting cost down.  Central control
is perceived as another benefit.

> Returning to the previous question, traditional systems where everything
> is intermingled on the on hand, and VMs providing total isolation and
> independance on the other hand, are the two extremes. In practice
> however you often want something in between.

There is not a single "you", of course.  Different applications demand
different features from the underlying platform.

Thanks,
Marcus