Something that’s come up recently in conversations with enterprises is the topic of building a cloud operating model. No, I don’t mean cloud architecture, nor do I mean cloud management per say, but a model to actually define and improve how you operate your clouds.
But, why bother with a cloud operating model, what’s the point? But more importantly, where is the value for the business?
Business Innovation – Driving Technology Complexity
As we know, technology is at the center of every business’s plans in today’s age.
In pharmaceuticals, artificial intelligence and machine learning was recently leveraged to develop the COVID vaccine. This would have been almost impossible to achieve as quickly without innovative and complex technology.
In finance, the world has moved beyond the desire to have an up to date, fancy smartphone app. Consumers are now starting to move their savings away from banks and into digital currencies like Bitcoin and Etherium (with winners and losers in that space recently!).
In retail, Amazon is a prime (pun intended) example of innovation through technology,with their massive use of AI and intelligent Robotics increasing the efficiency of their warehouses.
Businesses are finding more and more opportunities to leverage the latest and greatest tech to push them forward in the market and these very demanding business requirements have driven more and more complexity with I.T. services.
All this is happening with an intensity we have never seen before.
Specifically, we’re all leveraging cloud services more than ever. With these globally available marketplaces of software and data, available to use and customize instantly, businesses don’t need to start from scratch anymore.
These marketplaces of cloud services are available across various different clouds, for example, if you have skills in Lambda specifically, you will want to go to AWS. If you want BigQuery, you use Google Cloud.
Business and commercial decisions drive the consumption of different clouds too. Who do we already have a relationship with? Who has the best deal right now? Is this hyper-scaler a competitor of ours in another area?
All of this combined leads to a multi-cloud strategy for many, many businesses today.
Strategies for Multi Cloud
As you can imagine, getting to multi-cloud requires a lot of transformation across the I.T estate as a whole. Three things really need to happen to get there:
- Building a Cloud Strategy
- Planning Application Transformation
- Adopting a Cloud Operating Model
Building a Cloud Strategy
There’s a wide array of cloud services available from the many hyper-scalers and thousands of cloud providers, each cloud providing subtly different value. That value will inevitably change over time too, often. Different business and commercial requirements (cost, speed, risk) often drive the cloud strategy too.
Planning Application Transformation
Some existing applications will need to be migrated to the cloud, to extract more value for the business. Others might need to be modernized before they can take advantage of the cloud. Some of your applications/services could be directly replaced by a cloud service (databases are a common example of this). Essentially, any of the 5-8 R’s which are commonly talked about. Doing the right thing for each (business-serving) application is key.
Adopting a Cloud Operating Model
As you can imagine, having a cloud strategy and an application transformation strategy without having a model to operate in this new business-technology model can be a major challenge. How do you ensure costs stay under control? How do you scale out these services without breaking security or compliance rules? How do you integrate with existing tools and processes?
You’ll need a plan or a model focused at a higher level than features and functions of technology. The model needs to account for the people, processes and technology changes required to effectively evolve the organization into this space.
Building the Cloud Operating Model
There’s a few ways to approach building a cloud operating model, for example consulting services focus heavily on people and process, this can be a large engagement depending on the scale of the business. In this approach, organizational structures are reviewed and adjusted, new roles are defined and become accountable in this new world. Projects such as the introduction of DevOps teams are a great start.
Governance, operational processes and KPIs also need to be understood and accounted for too, for example, who can access the services? What approval policy and audit trail needs to be implemented?
Then you need the technology available to support those new initiatives too.
That’s one way.
The other way to do it is to lead with the technology, as we do at VMware, it’s our strength, right!
Multi Cloud Management Software
Multi-Cloud Automation software can encapsulate your clouds and provide an overarching API surface. This will increase the speed and agility of consuming, deploying and managing services across this new multi-cloud surface, driven by policy and with guardrails in place to keep things secure and compliant.
Transformational technology like this tends to lead to new operational ways of working, with DevOps teams born out of the increased lifecycle automation and new processes are created in policy, which can be applied continuously in the software.
Multi-Cloud Operations software can carve up your entire multi cloud surface into pools of resources available for lines of business to access. The software works to continually optimize the efficiency and performance of your cloud surface too. Bringing in software like this often births platform engineers of the future and S.R.E. type roles.
It’s increasingly important to look at Multi-Cloud Financial Management tools too, to keep costs optimized and predictable. Plugging into the billing APIs across your multiple clouds provides a level of control you’ll drastically need in this complex environment.
Leveraging this software, you’ll inevitably start building a cloud center of excellence as that higher order of control and management across the consumption of your clouds.
Bringing together your clouds and applying a multi-cloud operating model in this way gives you much more predictability and gives some staggering and tangible benefits:
Hours saved daily in provisioning and managing infrastructure and cloud services, enabling staff to focus on innovation and optimization, rather than feeding and watering what’s there.
Overall risk is reduced, with with a dramatic reduction in unplanned downtime and mean time to resolution.
When you optimize the cloud surface like this, you can also expect a reduction in software licenses, servers, storage and cloud costs because you’re not paying for what you don’t use.
However you go through this, I hope you see that thinking about and building a cloud operating model is an essential part of any cloud and application strategy
NB: This article was adapted from a video I recorded for YouTube, here.