Two Paths Converge in the Cloud: Realizing Value from Legacy Migrations and Native Cloud Development

istock 1160479733
PhonlamaiPhoto/istock

For enterprises built before or outside of the digital ecosystem, the pull of cloud computing can feel more like a shove. It needn’t seem that way.

The impact of the cloud

According to the TCS 2020 CIO Study, cloud computing has had the biggest impact of any technology on business. Three-quarters of CIOs said cloud’s impact had been “high” or even “extreme” in the past decade. And more than two-thirds (68%) continue to characterize cloud’s expected impact as high or extreme over the coming decade, joined by such cloud-dependent, cloud-enabling, or cloud-friendly technologies as artificial intelligence (79%), machine learning (72%), and the Internet of Things (67%).

Thus, the questions of just a few years ago – “why cloud?” and “when should we migrate?” – have been replaced by “how?” and “how quickly?” Yet while most leaders agree such migration is inevitable, there are skeptics whose doubts often are based in valid considerations or concerns. Advancements toward digitization and recent experiences of many enterprises with digital transformation provide ample evidence of the benefits and challenges corporations should be aware of when undergoing major technology migrations.

While the transformational benefits that the cloud promises are realized through sound and informed approaches, the optimal path for an individual company or organizational unit may still seem hazy initially – highlighting the importance of engaging early with a structured, efficient, and well-informed approach.

Two types of enterprises

There are cloud-native companies that were “born on the cloud” – typically the most recent entrants to the marketplace who are often found in business news headlines for disrupting industries and rewriting rules. Questions about migration rarely apply to these companies, as all their application development has been and will continue to be cloud-based.

However, for enterprises of longer standing – those making the paradigm shift to cloud from years or decades of development based upon on-premise and hosted infrastructure – whether and how to move legacy applications to a cloud environment, or for which purposes to develop new or completely redesigned processes built on cloud, often remain matters of internal debate and drive a constant refiguring of priorities.

In the TCS 2020 CIO Study of companies with at least $500 million in annual revenues, nearly two-thirds (64%) are at varying stages of digitization, according to their lead technology officers. Over half (54%) have completed proof-of-concept pilots and can claim some degree of enterprise-wide digitization.

Despite this digitization progress, 23% admit they still have a short-term focus and a lack of leadership when it comes to digitizing their operations, products, and services – while another 31% have strong leadership in place with initiatives across multiple functions and lines of business, and 11% even have new businesses, products, services, business models, or operating models as a result of digitization.

The opportunity offered by cloud for not-yet-cloud companies

For these traditional enterprises, cloud computing – the sheer availability of it combined with the benefits of increased agility, decreased ownership costs, and better positioning to compete and collaborate in the digital ecosystem – presents an opportunity to simultaneously modernize operations and innovate new lines of business. All of these companies, even those with an as-yet-minimal cloud presence, have years or decades of experience undergoing various technology (r)evolutions: mainframe, client-server, 3-tier architectures, enterprise resource planning, analytics – they’ve employed each of these and more as needed and have proven capabilities to adapt with shifting technology and market demands to drive business outcomes.

Corporate managers and boards know their firm needs to shed fat, be nimble, and always be adapting to maintain market leadership. They recognize generally the need to both modernize legacy resources and adopt cloud native at every opportunity, but often in conversations on these issues will become stuck looking at the macro-level view without supporting contextual details. This is after all what they are expected in their position to keep in mind:

  • What is the balance to achieve between migrating legacy operations and adopting the cloud as the native computing environment? Whichever they choose, each step will require investment – and to be rationalized by return on that investment
  • How best to optimize our operations? Every company knows this is a requirement but knowing which applications provide maximum value for the business can be a bit like reading tea leaves in the dark

New breeds and older legacies

Every established enterprise that’s been in business for 10 years and hopes to be for another 10 likely has both running legacy systems which day-to-day operations and quarterly revenue depend upon, as well as new ideas for addressing opportunities and challenges that could lead to new growth and innovation. 
As a general rule, anything that can be developed cloud-native should be. 
Companies may think they are saving money – and departments might be tempted by budget constraints -- to begin development on any available server/capacity, rationalizing that the small team making initial forays into a new project is enough resources to expend on untested ideas without committing to a cloud account/deployment, however nominal its initial charges. Such short-term thinking will only create technical debt when the time comes to port that initiative – or any initiative -- to the cloud so that it can scale for the enterprise and its clients.

Better instead would be to take advantage from the initial development stages of trends and advances in cloud computing – such as serverless microservices with polyglots, in which individual scripts can perform the same operations across multiple programming languages and applications to leverage the digital ecosystem, and/or containers, vendor-neutral environments for cloud applications that are more lightweight and portable than the virtual machines replicating whole servers.

TCS is today working with AWS – which runs 80% of all cloud containers currently in use – to harness the benefits of containers, microservices, and serverless technologies such as AWS Lambda for its clients. Technology is advancing at an exponentially increasing pace, and future changes are inevitable. Enterprises must change, and indeed must become nimble at changing, to adopt new capabilities in time to meet new market opportunities and challenges. Using the cloud as both the development and the runtime environment for new projects makes that kind of agility and flexibility possible.

But what about the legacy systems that must be migrated or otherwise mitigated? A number of approaches should be considered as at least one of them will be the best choice for any particular application based on the business requirements and outcomes desired.

Retain…or retire (and replace or rebuild)?

In all likelihood the decision to Retain an existing legacy system is a temporary one, deferring the decision of how to retire it to the future. When other concerns or priorities are more pressing or a specific horizon for deprecation or addressing the need is understood and planned, retainment may make the most sense.

The need to completely Retire old applications – either rebuild their current capabilities for a cloud environment or replace them entirely with a new cloud application for a new business environment – is usually apparent to those most reliant on the inputs and outputs of a legacy system. The “pain points” of current operations will help a business determine symptoms, causes, and a cure.

Rebuilding abandons the legacy system for a new cloud-based function or application that has been reimagined from the ground up. While the most initially complex and potentially expensive approach, for the parts of a business that themselves may need overhauling rebuilding/replacing may have a transformative effect as out-of-date and constraining systems are replaced with cloud native parallels that unlock the benefits of cloud. 

However, the seemingly limited choice of “retain or retire” are not the only options. Legacy migration is more of a spectrum of approaches than a typical binary decision.

Repurchase, rehost, replatform, or refactor/rearchitect?

Repurchasing (or, more commonly, relicensing) involves merely redirecting an existing application instance to use its cloud-based version, or retiring client-server on-premise/hosted software, in order to use a software-as-a-service version of a product instead.

Rehosting may be the fastest way for legacy apps to join their cloud-native relatives. Sometimes called “lift-and-shift,” rehosting merely moves an application or service from one physical server environment to a cloud environment such as AWS. You gain the reliability, availability, and efficiency of the cloud environment and utility costing model, plus future platform and application changes will likely be much easier than if taking place in your own data center with traditional infrastructure and operational considerations and constraints.

But given the past rate of change in technology, merely rehosting may very likely be forestalling the inevitable work of refactoring or even -- if too much more time passes and more technology iterations occur – the expense of rebuilding your legacy instantiations. Despite this question, rehosting may be the more prudent choice for a particular organization – especially if immediacy of availability is the primary consideration.

Replatforming is sometimes referred to as “lift, tinker, and shift”: It’s a variation on simple rehosting where a company takes advantage of the migration to cloud to modernize and implement new features in an application or perhaps swap proprietary components for open-source equivalents.

Refactoring legacy apps for the cloud is often the most successful approach when it is feasible. In refactoring the current runtime application, the company depends on is retained, but with modifications to some parts of it to make it cloud friendly. By moving to the cloud, further and future development can then take place more easily: essentially performing a rearchitecting over time. Refactoring involves rehosting – but more importantly, it lets you use the cloud as both a development platform – platform-as-a-service (PaaS) – and as a runtime environment (infrastructure as a service, or IaaS).

Rearchitecting, as compared to rehosting or refactoring, is a more ambitious undertaking – but it may offer the fastest ROI. The decision to rearchitect may be driven by an application’s or service’s current incompatibility with the cloud, but it may also be undertaken to lose no momentum in a company’s shift to ecosystem strategies.

In a Rearchitecting, the interoperating components of applications first get divided up as separate applications and microservices and are then remodeled individually in a cloud environment, thus emerging as cloud native when deployed – and thereafter useable by other enterprise apps and functions. Rearchitecting a legacy application instance could be seen as equivalent to remodeling a kitchen or even adding on a new wing while continuing to live in the house (compared to rebuilding, which is a complete start from the ground up).

Mindset may be the greatest defining factor in success

Technology and business are evolving quickly. Since the marriage of the two, this has always been true and, whenever it has been said it has never been truer than at that moment – including today. The gap between any enterprise’s cloud-native operations and its legacy, physically hosted applications will only continue to grow more quickly as cloud features and functionality continue to grow. Moving both new development and legacy upgrades to cloud environments can help decrease the size of the gap and slow the pace of its spread.

It is worth noting, however: This rapid evolution doesn’t present only challenges for an enterprise – the opportunities possible in a digital ecosystem are even more compelling for companies who choose to pursue them.

With every service that an enterprise might make available in a cloud environment like AWS, there is the potential to monetize that service. Large, established companies are the creators and bearers of great oceans of information. The market for that information – in aggregate, as outcomes and calculations, with the actual data protected and anonymized – is one of the fastest growing areas of new business development today.

Once a company is established in the digital ecosystem, many more efficiencies and advantages – from leveraging the API economy to employing analytics to create new business opportunities and operational improvements, to many other features of the ecosystem – become possible. Given the pace of technology change, hitching an enterprise’s growth to an established cloud platform and partner ecosystem can enable faster returns on investment, steadier revenue streams, and quicker times to market for customer-driven innovations.

Who’s already there?

According to the TCS 2020 CIO Study, almost a fifth (19%) of companies today are already thinking in terms of digital ecosystems when preparing for future competitors and seeking new opportunities.

1 2 Page 1
Page 1 of 2