One of the topics that have come up in FutureCIO discussions with technology, finance and business leaders in Asia is the issue of how much and how fast can organisations complete their digital transformation.
From comments, we can infer that there are little holding back businesses from embracing the cloud. Indeed, concerns and hesitations about moving the cloud have all but been dissipated following the start of the COVID-19 pandemic.
Yes, security and data protection are mainstay concerns. However, several operational issues remain in force. One of the recurring issues holding the transformation is application modernisation.
For Dr William Lee, research director for IDC’s Asia-Pacific (exclude Japan) Cloud Services group, the term application modernisation has evolved to take on a wide set of meanings.
He noted that enterprise needs for faster operations and being more flexible with responding to customers' and prospects' needs have accelerated priorities for application modernisation.
“Application modernisation refers to the refactoring and re-platforming of legacy applications to take advantage of the new enterprise digital infrastructure environment (in most cases, cloud-centric with either on-premises or off-premises as well as dedicated or shared),” he added.
On the topic of cloud migration, he referred to this as the migration of application workloads onto the cloud environment.
“It may or may not involve application modernisation in the case of lift and shift scenario. But in most cases, application modernisation is required at some point in the cloud migration journey to take full advantage of the cloud deployment model,” he elaborated.
Which comes first?
Lee says application modernisation is not necessary for cloud migration, although it is recommended to take full advantage of the cloud deployment model.
“Enterprises can already achieve some cost advantage through infrastructure modernisation whereby they consolidate and virtualized their infrastructure assets in their IT environment,” he opined.
He also commented that some legacy applications can simply lift and shift (rehosting) over to this new on-premises enterprise private cloud environment with zero to minimal architectural changes to the application.
“However, most legacy applications are monolithic in design and often not compatible with modern operating systems, hardware, and programming languages. This often inhibits the potential gain in new functionalities, speed of deployment and scalability of the deployment of these applications,” he continued.
Another incentive for application modernisation is the diminishing pool of talent able to support, maintain and manage legacy systems.
Lee opined that although application modernisation is not an absolute necessity, enterprises will still need to consider doing it at some point in time of their cloud migration journey.
Drivers of application modernisation
Lee cited three common drivers of application modernisation in the region. These include:
- Leverage on some of the cloud platform's functionality such as autoscaling and high availability
- Optimize the application functionalities for the cloud e.g., transforming monolithic applications to microservices architectures using containers
- Use of modern development tools such as cloud-based IDEs, DevOps toolchains, APIs, microservices and containers to extend the functionality of legacy applications
For Lee, enterprises see major public cloud platforms are becoming the number one source for innovation in new technologies like artificial intelligence (AI), machine learning (ML), big data, blockchain, Internet-of-Things (IoT), augmented reality/virtual reality (AR/VR), robotic process automation (RPA), serverless computing and quantum computing.
In addition, enterprises migrate systems to the cloud to acquire the scalability of public cloud infrastructure. He also listed out other benefits of cloud migration as:
- Movement of application development, testing, and staging operations to the public cloud environment
- Cost efficiency derived from IT infrastructure modernisation and consolidation
- Speed of deployment
Pandemic influence
Lee observed that when the unprecedented global COVID-19 pandemic hit, many organizations in Asia-Pacific had to fast-track their digital transformation journey to survive and grow in this new and challenging landscape.
He said this has resulted in increased investments in cloud services and technology, and the building of new capabilities. Cloud services can be deployed quickly with minimal initial setup costs, making them ideal as survival and recovery strategies for unforeseen crises like COVID-19.
What is in store in 2022
In the area of application modernisation, lee says IDC predicts greater cloud deployment, increasing numbers of production applications and highly interoperable workloads creates additional complexity.
He says: “legacy application modernisation has begun in earnest: 65% of developers in a worldwide IDC survey of developers noted that more than 50% of their legacy application portfolio had been modernized,”
IDC estimates that 51% of developers have rehosted a legacy application over the past two years. Meanwhile, 45% and 44% of developers have re-platformed and refactored a legacy application over the same period, respectively.
As regards cloud migration itself, Lee says cloud and cloud-centric operating models have grown to become integral components of the modern IT environment. IDC estimates that about 53% of Asia/Pacific (excluding Japan) already have production applications in the cloud environment (public and private cloud).
More than half of applications installed today is expected to move to a different environment over the next year.
By 2023, the private cloud will be the venue used for deploying 41% of all application workloads, compared with 31% for the public cloud - at the expense of all non-cloud methods.
Key success strategies in 2022
To ensure success, Lee suggests that enterprises seeking to modernise their applications conduct a modernisation assessment that provides guided insight into the level of effort required to modernize specific applications within a portfolio.
He recommended these companies consider upskilling existing professional resources to acquire relevant skills and partner with professional services firms where necessary.
“Assign a prioritization for modernisation of specific digital assets as well as a specification of the professional resources and skill sets that will undertake the requisite modernisation work,” he added. “Identify modernisation-related milestones that help assess the progress of their modernisation initiatives for specific digital assets.
On the topic of cloud migration, Lee said organizations should not make deployment decisions based on historical investments as it is important to consider all costs of maintaining any legacy systems, integrating them into new systems, and the cost to transition them to production.
He opined that all new applications and data workloads should go on a cloud platform of choice depending on requirements and availability, with a multi-cloud approach as the goal. Legacy applications and systems should be evaluated for modernisation for cloud deployment or retired if not in use.
He cautioned the importance of moving throughout the stages of a cloud strategy development in a quick and efficient manner. He advised organizations need a long-term vision when planning the migration.
“Understand what benefits may accrue to the business as it progresses from discrete cloud implementations to one that is optimized for cloud value. Then use those benefits to justify the appropriate investments in the future for your workloads,” he continued.
He also recommends organizations shift more applications and data workloads from on-premises to private and public clouds managed by third-party cloud service providers, legacy workloads that formerly reside in an environment that was entirely within the control of the enterprise must now evolve to include a management platform with cross-cloud policies and analytics to provide consistent security, compliance, and cross controls.
“Advancement in open standards and container technologies like Kubernetes has made modern application workloads more cloud infrastructure agnostic, offering true flexibility for current and future deployment requirements,” he remarked.
