[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Please let's not talk about "clouds"



On Mon, Apr 29, 2013 at 12:05 PM, Richard Stallman <rms@gnu.org> wrote:
>     These features have nothing to do with the way
>     you access the VMs.
>
> Thanks.  I understand a lot more now.
>
> Would the term "convenient management of VMs" fit these features?
> It seems that way.

Richard,

I think the end goal of all these frameworks, is to leverage a small
sysadmin/networking team to manage a large pool of resources, that
could include, but is not limited to:

1) Physical servers
2) Hypervisors
3) Networking equipment
4) OS installation
5) Block storage (drives)
6) DNS/DHCP
7) TFTP (and any other services required to support provisioning of
bare metal servers and VMs)

The end goal of this automation is to provide a self-service
infrastructure that does not require the involvement of this small
sysadmin/networking team, for allocating these resources, hence the
term "Infrastructure as a Service"

I'm not sure if this conversation has a place for "business" issues.
But I will share a few business drivers here, to give you some
context, and understand what is driving this trend.
1) Small highly technical team, can manage a very large infrastructure.
2) Resource management can be delegated to the end users of the
resources in question. e.g. - a sysadmin who knows how to manage an OS
doesn't have to know how to  build and deployment as IaaS
implementation. (The lower level stuff is all abstracted away).
3) Users of the service have a great deal of flexibility and can grab
resources on very short notice via simply running commands or calling
an API. Because of this, the end users also gain leverage as they can
specialize in using the API and IaaS infrastructures in question.

Overall, for infrastructures above the smallest scale, it's generally
more cost effective to run things this way, hence the trend in the
industry to move towards IaaS services. A side effect of this, because
the larger an infrastructure one is managing, the more cost savings
one can derive, increases the likelihood that "Public Cloud IaaS"
services, like Amazon Web Services, Microsoft Azure, Rackspace cloud,
and Google Compute Engine, will make up more and more of the
infrastructure people utilize for their server needs. Internal
computing that can not be run in outside infrastructures will also
move towards IaaS to remain competitive with external services.
(Consumers of server computing resources don't accept the whole 1-6
month acquisition cycle for servers any more.)

This is a trend that has been in play for close to 5 years now, and is
likely to only accelerate as these frameworks become more mature, and
the largest of these service providers are able to drive costs further
down through scaling to larger and larger sizes.

-Brian

> --
> Dr Richard Stallman
> President, Free Software Foundation
> 51 Franklin St
> Boston MA 02110
> USA
> www.fsf.org  www.gnu.org
> Skype: No way! That's nonfree (freedom-denying) software.
>   Use Ekiga or an ordinary phone call
>
>
> --
> To UNSUBSCRIBE, email to debian-cloud-request@lists.debian.org
> with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
> Archive: E1UWqZr-0000L4-FR@fencepost.gnu.org">http://lists.debian.org/E1UWqZr-0000L4-FR@fencepost.gnu.org
>


Reply to: