Five years ago, only the most open-minded companies would have used cloud to drive innovation as a core business strategy. Today, cloud is the mainstream choice for business-critical applications within the vast majority of enterprises, and the C-suite is certainly no stranger to the topic.
IT departments are now enjoying cheaper and more scalable services. While this is significant in its own right, importantly, it has set the stage for the cloud’s next phase as an enabler of business modernization. After all, the question now for many CIOs is not “should we move to the cloud?”, but “now that we’re in the cloud, how can we innovate?”
CIOs have a tall order ahead of them – under pressure to innovate at the same pace as industry disruption, all the while battling the complexities of hundreds of legacy applications, trying to keep the lights, and guaranteeing security and compliance. So what are the options available for CIOs innovating in the cloud, and moreover, how can they innovate efficiently?
Data on the cloud
Data is a good place to start. It’s at the heart of today’s business landscape – an ubiquitous resource that companies can leverage for innovations and insights. It’s all around us, and with the cloud, it can be mined very easily. Years ago, with highly structured data feeds, the process of collecting and analyzing data was laborious. Nowadays, provisioning is fast, and there’s a vast catalog of cloud services available to businesses which enables them to gain insights into data. These services, in turn, empower businesses to make smarter, faster, and more accurate decisions.
In the retail industry, for example, cloud is already a valuable instrument in determining event correlation. Businesses are able to feed in data relating to anything, from sales, to IoT devices, to the weather, to market sentiment – all of which can be aggregated and analyses in order to provide better understanding about customers and their needs.
When combined with machine learning and AI, further innovation is possible. AI, of course, has been around for a while, but it used to be expensive and complex if done on-premise. Now that barrier to entry has vanished, with out-of-the-box offerings from most public clouds, customers are able to attain quicker return on investments with this technology. This is shaping up to be the next battleground for public cloud providers and each are making significant investments in this space – all of which ends up benefiting customers. The technology is also becoming more simple meaning you no longer need PHD level data scientists to gain unique insights into your data.
The real power, of course, comes with the advent of Real Time AI. This technology will change the way we work and create scenarios where AI not only responds to low stock, but can predict low stock and manage stock levels on your behalf. This type of technology has the potential to disrupt industries, supply chains and even the economy as we know it. As such we must tread carefully and be deliberate in its application. Personally, I see AI as the next step in increasing our efficiency and accuracy just as the spreadsheet did for the accounting and finance industry, which enabled those professionals to focus on more value-added activities.
AI doesn’t only exist in the cloud, but also at the edge. A brilliant example of this is in the truck haulage industry. It is my belief that in the foreseeable future we will still have a driver in the cabin who is ultimately responsible for the vehicle, its contents and the people around them. Nonetheless the technology provided will vastly increase the efficiency and safety of the vehicle resulting in positive outcomes for the business, the environment and society in general.
Out with the old, in with the new
On the one hand, cloud is a force for innovation because it becomes the bedrock for creating new things: the cloud enables projects to be spun up quickly, which allows businesses to prototype rapidly and learn quickly from what doesn’t work. Combined with agile processes, organizations can quickly build working prototypes. All in all, cloud can be an agent for new architectures and processes.
But it’s easy to forget that cloud is also an innovator for ‘the old’: the cloud presents businesses with the opportunity to re-evaluate their application portfolio and retire, consolidate, and resize many services and applications. Very rarely do organizations manage their application portfolio in a holistic manner, leading to duplication of capability and services across the organization. This is further exacerbated when organisations are acquisitive; in these scenarios only the most basic plumbing is put in place between the organizations and the more complex system integrations put into the ‘too hard to do’ basket. This creates technical debt, which can lead to a downward spiral in organizational delivery capability.
The services that are to remain will fall into one of three categories; those that can be migrated into SaaS solutions, those that are built in-house and can be re-factored into a cloud native application, and those that will need to remain on IaaS for the foreseeable future.
The cloud providers have created a myriad of technologies and capabilities to deliver applications and services, to the point of creating ten different ways to skin the cat! The latest kid of the block that can be used to drive innovation (and confuse us all) is containers. Containers are an enabler for microservices architecture, but the core value here is ease of replication between different platforms: if it runs on a software engineer’s laptop, it will run in production on one cloud just as perfectly as it will on the next. This speeds up the time-to-live and mitigates deployment challenges (e.g it worked fine on my laptop).
The alternative option is serverless, which would be a great use case for net-new applications and services and so isn’t quite an all-around substitute for deploying and managing your own containers. Nevertheless, it serves its own purpose. When it comes to serverless, simplicity is king: an engineer simply writes some code, sets a few configuration parameters, and uploads everything and voila, they are up and running. Serverless is designed to be event-driven and loosely coupled so the engineer can quickly stitch services and components together to make an application.
Inevitably, there are applications, either commercial or home grown, that won’t have SaaS or cloud native equivalents. Maybe there’s an application that’s not worth the investment to change the codebase; maybe the vendor themselves haven’t made it available for the cloud; maybe it’s been so heavily edited that the company just can’t migrate it to the cloud because it won’t have all the same features. There’s a number of reasons why refactoring or SaaS may simply not be possible, and these kinds of applications can be run on virtual machines.
Here, the company can start to innovate with capabilities like Amazon Relational Database Service or Azure SQL Database Managed Instance. These technologies enable businesses to simplify application components by consuming these as services rather than servers – leading to fewer things to manage and lower costs.
Choosing to innovate, and choosing what to innovate
Once companies have adopted cloud computing, they have a number of innovation avenues to take – but choices need to be made. Larger enterprises, in particular, need to be targeted in what they choose to innovate because there’s only so many hours in the day and so many resources at their disposal. Some organizations fall into the trap of planning to convert everything they run into PaaS, but the effort and the expense of this often offers a poor return on investment, at least in the immediate future.
Ultimately, if hard choices have to be made and push comes to shove, the investment and the hours should be prioritized on the items which are closest to customers. This has the most impact on the bottom line.