Freigeben über


If we start with the premise that Automation is a Good Thing …

I spend a lot of time thinking about the role that “Automation” plays in various scenarios.  What does this look like in todays Datacenter, what does it look like as we move to Private Cloud Architectures, and what role will it play in situations where both Private and Public cloud architectures are in play?

I think before we start this discussion, there are some definitions, some paramterization if you will, as to what I mean when I say “Automation”.

Automation to me is actually 3 things.  It’s Integration, It’s Orchestration and It’s Automation. In that order. Makes perfect sense right? Where I come from, the response would be Yeah Right.

So, let’s work backwards through this. If we start with the premise that Automation is a “Good Thing” and something that we desire, or should desire (because everyone should want repeatable, resilient and compliant processes) then we need a path to get there from where we are today.

One step back is Orchestration.  Before we can Automate a process, we need to be able to link together the discrete tasks that make up the process, and have them occur in a timely and consistent fashion. Like the conductor of a set of musicians, you need all the instruments to be played in the right way, with the right notes in the right order.  This is Orchestration.

And before we can Orchestrate, we need to be Integrated.  If you can’t reference your technology environment  as a single entity, with all the different systems from different vendors connected together, then you have no way to Orchestrate and therefore any Automation you attempt will at best deliver poor results.

Integration, followed by Orchestration and rounded out through Automation.  You know what you have, it’s able to be thought of as an entity, you can move all the parts together and in an orderly fashion.

Today, very few Datacenters can make a valid claim to have these capabilities.  It requires standardization, it needs the removal of individual component dependencies from their underlying platforms, means you must have mapped out and understood exactly how your business is mapped against the technology you use to run it, and it requires a set of technologies that are only just starting to emerge and mature.

We have made no secret of the fact that the core tenancies of our Private Cloud are derivatives of our learning's from running some of the largest Public Cloud systems.  You think we have any non-integrated, non-orchestrated, non-automated parts to Bing, Live, Xbox and Azure?  You think when you log in, and in a self service orientated fashion provision a new set of services that this is anything other than an on-demand automated process?

Sure, your environment is a lot smaller than these massive environments. And when I say massive, these are simply large on a scale that can be hard to comprehend.  But you want all the benefits from these learning’s in a scaled down manner so you can run an efficient, dynamic company.

Which leads me to the question of what this looks like in todays Datacenter, and what it will look like in the Private Cloud?

Today, the best thing you can do at this point in time is start mapping out how your business uses technology, what the processes are, what the approval mechanisms are, what your compliance and regulatory requirements are, and how you intend on rationalizing your Datacenter into a Private Cloud (and/or what your adoption of Public Cloud can/will be).  This is a significant piece of work, and will lay the foundation for tomorrow.

As you move from todays OS + Application stack into a dynamic platform with composed applications, service templates and an SLA driven self-service user centric computing model, you will need deep application insight into how your services are performing, you will need to be able to reference your physical infrastructure as a cohesive fabric, your compute resources as delegated clouds, a knowledge repository that captures all your critical business information and processes, and you will need a way to recover in the event that something goes horribly wrong.

And then lets look at what happens when you start bursting to the greater Public Cloud, when you move specific services such as email to external shared service platforms and your workforce continues to be more and more mobile and use a greater range of devices from every vendor on the planet.

Suddenly, that early reference point of understanding exactly how your business uses technology to run becomes the most important thing you have.  Because if you have that, then you can cope with all this change, with this generational shift in computing dynamics.

And the role of “Automation”, as defined above as Integration, Orchestration and Automation?  Well, that’s what makes it all work.  That’s what is going to make the transition possible. And that’s what will let you sleep at night.

It’s going to be an epic next few years.

A.