Automation Logic were delighted to present with our Client, David Rogers from the Ministry of Justice, this week at the Central Government Business and Technology event (CGBT). CGBT is the UK’s leading event dedicated to sharing best practice, emerging trends and innovations across the Civil Service.

MOJ have been an early adopter of public cloud technology within government and have partnered with us at Automation Logic to build and operate the cloud platform on which they run their new digital services. To learn more about our work with the MOJ, check out our case study.

Our CGBT presentation Hybrid Clouds. How to go slow and haemorrhage money doing it” centred around expelling the false promises associated with hybrid-cloud.  In the first of a three part blog, we’ll explore these themes in more detail.  We’ll introduce the two main types of hybrid cloud, and where the interest in hybrid cloud stems from.  We will then expose the reality of hybrid cloud to be a combination of mis-marketing and over-engineering, and show how a much simpler strategy (which we call ‘multi-cloud’) conveys all the benefits that hybrid cloud fails to deliver, and does so with greater speed and in a shorter time.  We’ll conclude by describing how a  collaborative, multi-cloud strategy is working really well for Automation Logic and the MOJ.

Our first type of Hybrid Cloud which we call ‘Private/Public’ can be defined as a mix of private (i.e. on premises) and public cloud hosted resources, combined and consumed as a single, unified service.

Many organisations opt for this type of hybrid cloud believing it will enable them to support different workloads depending on factors such as data sensitivity, data sovereignty, compute architecture, service architecture.

Our experience, having successfully delivered cloud engagements to clients across Central Government, Banking, Retail and beyond, tell us that the arguments for hybrid cloud just don’t stack up.

For Private/Public hybrid clouds, where the Private side is based on existing infrastructure*, it is almost never worthy of the term cloud.  Not to say these on premises systems are not genuinely useful, they are, but they are typically only virtual machine provisioning platforms, albeit with some advanced automation.  They are almost always missing several of the key characteristics that would warrant the term ‘cloud’, e.g. usage based billing, massive elasticity, limitless scalability or direct API access. In practical terms, you can’t spin up 1000 machines in 5 mins, destroy them and only pay for what you’ve used.  Whilst it may be unfair to hold these systems up to these kind of standards (they were never designed for that) it’s also disingenuous to call them clouds.

So what you really have with a Private/Public hybrid cloud is public cloud attached to something that you’re calling private cloud, but it is not really a cloud at all….

Why is this important, after all, what’s in a name?

Apart from the disappointment of buying into a set of expectations and finding yourself short changed, adopting hybrid cloud as a strategy sets a precedence within your organisation that there is parity of capability between the Private and Public elements when, as we have just discussed, that is rarely the case (particularly if we’re talking about private infrastructure that is more than a few years old).  

Framed in this way, hybrid cloud as a mix of aging (not very cloudy at all) infrastructure and public cloud.  Well, that’s not a strategy, that’s a half-finished transformation.

Hybrid Cloud is not a strategy, it’s a predicament.


An argument for hybrid cloud is an argument against greater public cloud adoption and a drag on your transformation to modern hosting and digital services.

In the next part of this blog series, we’ll examine the other type of hybrid cloud that we frequently encounter: the broker or abstraction layer.  We’ll also introduce the possibility of a simpler way to attain the benefits that hybrid cloud fails to deliver, an approach we call multi-cloud, and how this is working well at the MOJ.

Today’s blog was written by Automation Logic Co-Founder Kris Saxton.

To discover more about Multi-Cloud, read on or get in touch.

Contact us today –

*Although there are still some edge cases for private clouds, anyone (is there anyone?) seriously considering building private clouds in 2017 must have *very* strong business reasons for doing it.  Reasons that would trump the higher costs, lower operational resilience and weaker security that come with running your own  infrastructure without a multi-billion dollar company at your back.

Automation Logic’s purpose is to deliver an automated world where people are free to realise their creative potential. We deliver technology-enabled transformation, helping our customers adopt emerging business practices and IT. Our portfolio of services span consulting, implementation and operational management solutions for DevOps, Cloud and Automation.

< Back