Why are you going to the cloud?

OK, ok. You are already in the cloud.

What I mean is why are you really going to the cloud?

You are an enterprise with a generally complex technology landscape. It runs across many data centres and clouds. It is an estate of IT applications built on all sorts of technology stacks built using mature skills and experience. They are maintained by people that are still employed by you, or via outsource, or maybe not at all because you’ve lost the knowledge. You can get a head start on understanding the IT application estate from your long-term relationship with technology and service suppliers, such as IBM, who can help companies “born in the enterprise” tip to “running on the cloud”.

The complexity may well be virtualised and be automated to a mature level. However, the majority of the estate is not built and run according to a modern cloud architecture.

But you are already in the cloud and are seeing its value.

With over 25% of IT estates already moved to the cloud, we are building up to the tipping point. Really being in the cloud means the IT operating model needs a step change and operating in the cloud is no longer an exception.

Cloud as an operating model means more strangers are going to be doing things relating to your IT application than you used to do.

Cloud as a technology model means the IT applications that serve your business need to work with platform services that automatically change (for the better) under you.

In 2011, the NIST definition stated “cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g. networks, servers, storage, applications and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.”

NIST further described “five essential characteristics of cloud computing: on-demand self-service, broad network access, resource pooling, rapid elasticity or expansion, and measured service. That computing is delivered through three different service models (infrastructure, platform and software) and deployed in one of four ways (private, community, public and hybrid).”

Fast forward to today, the cloud computing model dominates the IT industry through combinations of these characteristics and has created a significant ecosystem around it. The has been an explosion of open source projects, whose code is now embedded in mission critical global scale cloud platforms. People and skills have changed; agile software delivery processes are on the increase within IT departments (and the outsourced service providers). Adaptable programme deliveries, experimental organisational design and “ship it” / “fail fast” mindsets are infiltrating seasoned and regulated corporations.

As a software engineer creating and maintaining a business service, I am going to be able to deliver that system in a more effective manner because of the cloud. Cloud gives me a totally software-defined way (“no humans involved”) of deploying an IT system. I am in control with less dependencies on others and I can see results from code quicker.

The public cloud service provider (“somebody else”) has solved some wicked problems. They have been the system integrator of compute, network, storage and software frameworks. They have solved security and they keep pace with the cadence of open source releases. They have solved this at global scale and providing access virtually anywhere.

This has in turn enabled enterprises to build and operate cloud platform functionality in data centres under their control. Private clouds become necessary when particular workloads have non-functional requirements that cannot be satisfied by the public cloud service provider. This could be due to high performance compute demands, unavailability of the public provider in a region, regulatory data constraints and connectivity to traditional enterprise IT systems.

Public or private, software engineers can embed software-defined policies that can reliably land their application in the right physical location.

Enterprise best realise the benefit of cloud by adopting a hybrid cloud strategy. This is one where there is a single, integrated approach to building and operating cloud applications on-premise, on private and on multiple public clouds. Such a strategy enables a responsible, transformative evolution of the IT application estate to new cloud operating model ways of working and benefits.

An internal IBM study in conjunction with a leading consultancy has analysed the benefits of a hybrid cloud strategy. It is used to assess companies cloud transformations and shows benefits such as an increase in the number of applications migrated to the cloud model, duplicative tools and processes removed, lower costs in certification, and reduced cybersecurity and regulatory risks.

For an illustrative large enterprise client with $1B of IT spend, a hybrid cloud approach can realize $200 — $270M of annualised value; 2.5x more than public-cloud-only approach.

The result is that IT applications built to operate and take advantage of cloud development practices and platforms are more responsible enterprise IT assets.

DevOps delivery lifecycles (and their measurements) with consumption-based pricing gives me transparency of costs. The software and platforms I use to build and deploy the system onto scale and are elastic — there is an element of “for free” future-proofing built in as usage increases. The global topologies of cloud service providers with availability zones and synchronous self-replicating data stores offer higher levels of availability and resilience over traditional on-premise IT infrastructures.

Understanding why you need to shift your business to the cloud helps inform both the design of your hybrid cloud operating model and target architecture and technology selections.

Understanding why is across three broad dimensions: (1) cost; (2) risk; (3) and agility.

Cost. The total cost of ownership of solution should be more transparent and granular. It brings more certainty to business cases. The cost of getting there may be less transparent. Quite often legacy IT operations are not able to offer granular costs pertaining to “hosting an application”. With a myriad of interfaces and shared componentry it is often difficult to just turn one thing off and move it to the cloud.

Risk. The nature of well-engineered, secure, global scale cloud platforms brings massive benefits to the levels of service that can underpin an IT system, thereby reducing its risk profile. However, the control and integrating of a cloud provider into the operating model is not a trivial task.

Agility. The speed of delivery and access to a modern catalogue of software features, that are kept current and at pace is a huge agility lever. The wide variety of services and things that can made possible with open source technologies is mesmerising. However, building something cool and innovative quickly shifts to maintenance, tracking the provenance of building blocks and their maturity.

I believe the biggest reason why you are going to the cloud is for agility.

The innovation available through cloud as the foundation of an IT system enables a business to be agile and gain competitive advantage. Agility to test better features and experience that can show differentiation rather than competing on price, helping focus on increasing revenue.

The simple act of modernising an existing application to automate its deployment and be securely deployed and operated in the cloud is an incremental innovation that brings increased agility to the business application owner.

As enterprise technologists we now have to engineer and integrate multiple clouds to help bring that agility at scale across the whole business. It is now our time to help the business understand why they are going to the cloud and streak past the tipping point successfully.