Before we examine Consolidation, the next step in the journey to
cloud, let's take a short detour to address a critical choice you
will face at the outset of your journey: whether to deploy your
databases in virtual machines or not.
A common misconception we've encountered is the belief that moving
to cloud computing can be accomplished by simply hosting one's
current operating environment as-is within virtual machines, and
then stacking those VMs together in a consolidated environment.
This solution is often described as "Infrastructure as a Service"
(IaaS) because the building block for deployments is a VM, which
behaves like a full complement of infrastructure. This approach is
easy to understand and may feel like a good first step, but it won't
take your databases very far in the journey to cloud computing. In
fact, if you follow the IaaS fork in the road, your journey will end
quickly, without realizing the full benefits of cloud computing.
The better option to is to rationalize the deployment stack so that
VMs are needed only for exceptional cases. By settling on a standard operating system and
patch level, you create an infrastructure that potentially all of your databases
can share. Now, the building block will be database instances or
possibly schemas within databases. These components are the
platforms on which you will deploy workloads, hence this is known as
"Platform as a Service" (PaaS).
PaaS opens the door to higher degrees of consolidation than IaaS,
because with PaaS you will not need to accommodate the footprint (operating
system, hypervisor, processes, ...) that each VM brings with it.
You will also reduce your maintenance overheard if you move forward
without the VMs and their O/Ses to patch and monitor. So while IaaS
simply shuffles complex and varied environments into VMs, PaaS
actually reduces complexity by rationalizing to the small possible
set of components.
Now we're ready to look at the consolidation options that PaaS
provides -- in our next blog posting.
