Companies can achieve huge cost, operations and productivity benefits from migrating their data and digital processes to the cloud. It takes careful planning, management and execution, along with some forward-thinking, to protect organizations from headachesâ€”and higher costs than they had anticipated.
The costs come in a variety of flavours: sticker shock, technology lock-in, vendor lock-in, failed migrations, controlling access and so on. This could have a significant impact on companies during the entire lifecycle of the application being deployed in the cloud.
Data is the lifeblood of modern business, and managing it properly is essential to being able to extract all the value it offers. The need to manage, cleanse, transform, protect and get value out of data rapidly makes all the difference. Companies that master their data can dominate their competition.
Traditionally, companies have managed their data in-house, using bought-in or licensed technology and software. This is fraught with risk. What the pandemic taught us is that this model does not work; when resources are unable or unwilling to work in the data centres, operations slow down and impact the business. The hardware on which these applications run become obsolete and unwieldy when it comes to scaling the operations to meet higher business demands. The provisioning of additional hardware is a time-intensive and expensive process.
Cloud computing comes as a saviour to those companies that have been stuck with the traditional on-premises approach to enterprise computing. The data centres are all virtual, and cloud providers offer SOC 2 compliance. The principles of SOC 2Â are security, availability, processing integrity, confidentiality and privacy. Scaling on-demand to address peak computational demands, point-and-click provisioning of new hardware and high-speed storage that is configurable all make it worthwhile.
The benefits attributed above have triggered a significant exodus to the cloud. As a result of that frenzy to catch up, little attention is paid to what is moved and how itâ€™s moved, opting for expediency over the right way to do it. Companies have embarked on expensive cloud migration projects without putting proper thought into them. The net result is delayed projects and expensive failures.
While companies continue to move applications to the cloud, they may never be able to get 100% of their applications moved to the cloud. The norm will be whatâ€™s commonly referred to as the hybrid multi-cloud environment (i.e., applications distributed between on-premises hardware and one or more cloud providers). As the marketplace evolves, applications may move back on-premises or between cloud providers for a variety of reasons, including cost and innovation. The added complexity in this new world of computing needs companies to pay close attention to aspects such as computing costs, ease of use, security, compliance and privacy. The inability to get insights can result in higher operational risk exposure, loss of reputation and regulatory fines.
Lift And Shift
This is a model that is very popular and considered expedient. The reason itâ€™s expedient is that the original application is either converted as-is into a new application using automated scripting or manually converted to the cloud within the same architecture as it existed on legacy systems. But this can result in duplicated, corrupted, redundant and orphaned data that caused problems on-premises to pose the same challenges in the cloud. The only exception is that it is now costing the enterprise by the hour. The net result is significant cost overruns on a monthly basis.
Replatforming involves transporting the application from the on-premises hardware directly to the cloud. The challenges involved here are in making sure that the connectivity between existing applications and the re-platformed application is maintained and the security of the information exchange is preserved. If done without proper knowledge of the interaction between applications, this could result in missing data, an expensive search and rescue effort to reconnect the missing applications and security and compliance risks, in addition to carrying inherent inefficiencies in the application into the cloud and causing significant cost overruns.
This model involves rewriting legacy applications to leverage cloud-native solutions. There are pros and cons to this approach. Going cloud-native may result in the introduction of new technologies that could require the retraining of resources. The pros could include better performance and availability of resources in the market. The cons, however, could be platform and vendor lock-in, which would make it very difficult to migrate out of the cloud provider should the costs or offerings become unattractive for the enterprise over time. This will most likely end up in additional costs to bring the application back on-premises or move it into another cloud provider’s environment.
Doing It Right
Planning: Companies need to weigh the advantages of moving applications into the cloud versus the additional responsibilities that they incur because of the movement.
Agility: Most of the migrations happen while the existing applications may continue to evolve. Getting a near real-time picture of the migration and having the ability to compare the before-and-after picture to make sure that there is no loss of business continuity, added risk exposure and noncompliance resulting from the migration effort becomes critical.
Continuous Monitoring And Fine-Tuning: While any of the techniques mentioned aboveâ€”such as lift and shift or re-platformingâ€”can be adopted, companies need to continue to pay attention to the application and iteratively fix aspects such as code duplication, code quality and security that could adversely affect the benefits of the cloud. This can help avoid the shock of bloated monthly bills.
Long-Term Cost Benefits
Too many companies have found themselves in as much trouble after cloud migration as they were before, simply because the process wasn’t planned, managed or executed properly. Getting that data back into shape will be expensive.
Hiring a third party to manage and execute the migration needs careful budgeting, but itâ€™s important that data managers stress the company’s longer-term imperatives when making a business case for such a move; the overall savings that a properly managed and trouble-free migration offers will easily outweigh those of a badly executed in-house attempt.