Michael Allen, VP and EMEA CTO, Dynatrace, discusses the steps to achieve successful Digital Transformation on a journey to the cloud.
There’s no doubt that everything is moving to the cloud. The International Data Corporation (IDC) forecasts that global revenues from cloud services will reach US$554 billion by 2021. However, while we are all undoubtedly aware that the continued drive to the cloud is accelerating, many organisations are faced with the challenge of how to successfully migrate their applications to the cloud.
For businesses to fully benefit from the efficiency and agility benefits that the cloud has to offer, it has to be more than a simple ‘lift and shift’ operation. Organisations that just move their existing applications directly from an on-premises environment to the cloud will struggle with inefficient resource consumption and cumbersome code that makes it difficult to optimise performance and innovate at speed. However, re-building everything from scratch can be incredibly costly and time-consuming. That leaves the question; how can organisations increase their chances of ensuring successful transformation in the cloud?
Charting a course to the cloud
Before any decisions are made about how they are going to migrate applications to the cloud, organisations should start with an assessment of their on-premises environment. They need to identify which applications should be moved to the cloud, which should remain on-premises due to security and compliance requirements, and which can just be scrapped, or replaced with a Software-as-a-Service (SaaS) solution to simplify migration and ensure the best possible user experience.
For custom-built applications that are to be migrated from on-premise to the cloud, often the best approach is to re-architect them into microservices and containers. These cloud-native architectures make it easier for organisations to fully reap the benefits of the cloud. They allow organisations to achieve the agility and flexibility that means application resources can scale in line with fluctuations in traffic, operational and storage costs can be reduced, and faster implementation and deployment can be achieved.
Storm clouds ahead
However, while the eventual benefits of rearchitecting applications far outweigh those of the ‘lift and shift’ approach, businesses can be put off given the tricky and costly processes of redesigning, recoding and repurposing applications; not to mention the added complexity it brings to the enterprise cloud ecosystem. Research has found that today, web and mobile transactions already cross an average of 35 different technology systems, compared to 22 just five years ago. In large part, that’s because organisations are running hugely complex hybrid cloud environments, with applications that sit astride multiple cloud platforms and on-premises systems.
Microservices and containers add even more layers of complexity to that technology stack, while creating black box environments that are difficult to see into. These environments are also more dynamic, meaning that applications that were previously static are undergoing constant change. The extra ‘noise’ this creates can make it infinitely more difficult for businesses to identify and rectify the root cause of any performance problems that arise, putting user experiences at risk.
Combatting the complexity conundrum
As a result, before they begin the process of rearchitecting applications for the cloud, enterprises must undertake analysis of where to optimally break the monolith into microservices to ensure they will actually work in this new environment and deliver the desired benefits. They must also ensure they have the ability to maintain end-to-end visibility and control over the user-experience in spite of the added complexity that’s being introduced to their hybrid, multi-cloud environments. The first step in that process is to identify a way to ensure rearchitected applications provide the same level of performance and quality of user experience when they’re in the cloud as they did on-premise.
Baselining application performance before migration will provide a clear picture of what ‘normal’ behaviour looks like, so organisations can quickly identify any degradations introduced by the process of rearchitecting applications into microservices and containers. Having established that baseline, businesses need a way to monitor the performance of their newly rebuilt cloud-native applications in real-time, so any problems that arise can be resolved before the user experience is impacted. Traditional monitoring approaches are simply unable to provide that level of visibility in today’s complex cloud-native ecosystems, which introduces a major risk, as application performance can’t be guaranteed – a risk no business can afford to take.
Clear skies ahead
This challenge can only be overcome with new monitoring approaches that have advanced AI and automation at their core. These capabilities mean that IT teams can auto-discover dynamic IT ecosystems and instantly analyse the complex interdependencies between cloud-native application and infrastructure components. They can also auto-adjust the performance baseline as the cloud ecosystem changes, to reduce alert noise that stems from the fluctuations in resource consumption that are common in the cloud. This provides the business with real-time situational awareness into their cloud-native ecosystems, making it much easier to identify and rectify performance problems before they impact user experience.
As with any major technological shift, rearchitecting applications into microservices and containers is not without its pitfalls. However, with the right measures in place to ensure they can steer a steady course through the complexity that these cloud-native architectures bring, businesses will be able to fully reap the rewards of versatility, agility and flexibility that the modern enterprise cloud has to offer. This capability will be a major advantage in the race for constant transformation that’s taking place across all industries and every organisation.
