If you’re embarking on a move to the cloud, you may have some misconceptions about the benefits of a cloud migration project. Here are five considerations before forging ahead.
It’s widely acknowledged that cloud technologies have become critical to modern IT services and service delivery.
At this point, most organizations have at least evaluated cloud computing architecture and how cloud models might affect their businesses. Others have reached advanced stages of cloud migration and transformation. Moving to the cloud invariably introduces many new business challenges and issues that need to be addressed for successful migration.
Here are five common issues that organizations need to consider before they can drive successful transition to the cloud—whether they are just starting the journey to the cloud or well entrenched in their transformation.
One of the most common misconceptions about the cloud is that it automatically and immediately reduces operational costs and provides return on investment. The reality is most organizations will spend more money in the first several years of a cloud migration than they would if systems remained on-premises. Further, some applications may cost more to operate in the cloud for the life of the application than if the app remained on-premises.
There are several reasons the cloud costs more at the outset.
As companies consider cloud migration projects, understanding the true costs of on-premises operations will help in planning and considering cloud vendors. IT organizations that recognize the transition to the cloud will cost more than operating on-premises will be better prepared.
Containers have become an important tool in cloud migration projects as well by enabling IT departments to deploy applications more quickly and without launching an entire virtual machine (VM) for each app. Instead, multiple isolated systems, called containers, run on a single host. Ultimately, these efficiencies can help create faster, more flexible and innovative product development.
There are two major forms of containerization today: (1) containers that act and behave like traditional VMs that abstract the operating system kernel and (2) containers that abstract the underlying infrastructure and provide application programming interfaces (APIs) and development tools for building apps that sit on the abstracted infrastructure.
In the case of some container technologies, an app can be built once and run on-premises or in the cloud without requiring changes. On the surface, this approach provides the greatest flexibility for building and running apps and avoids lock-in with a single cloud platform. In a world where 67% of enterprises have a multicloud approach and 90% are expected to have one by 2019, this container strategy makes sense.
But the decision to use containers may bring unanticipated impact. In both kinds of containerization, the app doesn’t exploit native cloud capabilities and it may not fully exploit the global scaling options available in cloud platforms. Further, in both cases there is still underlying infrastructure that must be managed and configured. While major cloud providers have tried to simplify container management, there are limitations, and organizations may require infrastructure management negating some of the value of the cloud. Further, these container technologies are not true serverless compute platforms.
Before planning the use of a specific containerization approach, consider what is most important in the use of the technology. If the goal is to have the greatest portability, container technologies are a solid choice. If vendor lock-in is less of a concern and the power of elasticity, serverless compute, and global scale are important, containers may not make sense.
While cloud computing may address many business needs, some applications require very low latency and more immediate processing. A machine on a factory floor that begins to malfunction, for example, can’t wait for data to be collected, parsed, sent to the cloud, have complex event processing rules applied to the data, identify the anomalies happening, then generate the appropriate alert; this round trip to the cloud is time-consuming. In some cases, this delay could be the difference in major events in machine operations or even loss of life.
With edge computing, cloud capabilities are brought on-premises, and these capabilities—including complex event processing and machine learning rules—are processed on-premises and close to the users, devices and data being used.
Prior to making the decision to migrate to a cloud platform, consider the available options for edge computing and how those capabilities will work and affect the business. This evaluation is especially important for applications where Internet of Things- (IoT)-connected devices will be integrated into business processes.
Organizations often begin a cloud migration project by identifying the easiest apps to migrate to the cloud. While this approach works well, one of the most important decision points in identifying the apps and services that can and should be moved to the cloud revolve around the idea of data gravity.
The underlying issue with data gravity is that most business apps and services that depend on data will operate in the environment or cloud in which the data resides. Apps that migrate to the cloud that depend on on-premises data often suffer from network latency issues. Likewise, apps that remain on-premises where the data resides in the cloud may also suffer from network latency. In both cases, some business apps won’t function correctly and may even cease to function when these network issues are introduced.
Data gravity issues become more complicated when a major source of business data is moved to the cloud, organizations need to plan on moving numerous applications and services at or near the same time the data is moved to minimize the business interruption and potential application issues. Enterprises make the mistake of failing to identify apps and services that depend on on-premises data. Many organizations operate a configuration management database (CMDB) solution, but many CMDB services aren’t complete and updated. In these situations, real-time analysis tools are needed to identify applications and services using business data to plan and understand data gravity issues.
Many organizations migrating to the cloud do so because executives have mandated the transition or organizations see competitors making the transformation. Further, many organizations begin a cloud migration project with the intent to save money. In general, these organizations will be the least successful in their cloud transformation. The most successful companies embrace a journey to the cloud to drive digital transformation and innovation. The most successful companies make major business changes and use the cloud to power that transformation.
As cloud platforms provide pre-built services, and on-demand scalable services, organizations can focus more time and effort on creating business value. And these services can use advanced capabilities, such as machine learning and artificial intelligence in the cloud to generate new business solutions. Examples include companies that historically sold a piece of equipment transforming into a new equipment-as-a-service business models where predictive maintenance and analytical insights can now be sold as business services. Other companies now create entirely new services and industries.
And “born in the cloud” companies are creating significant disruption and transformation across many industries. The cloud is about making a broad range of services and capabilities readily available to solve business challenges and create new offerings. Organizations that approach the cloud with this focus of innovation and digital transformation are more likely to be successful in the cloud.
Whether your enterprise is embarking on the cloud journey or is in more advanced stages, application portfolio management can help avoid data gravity issues while creating a complete picture of the cost of operating and running an application on-premises. A complete application portfolio can help companies drive better business decisions on the financial impact of the cloud and which applications are worth migrating to the cloud. Regardless of how well an app might fit in the cloud, consider the question of how moving that app or building that app in the cloud will drive business transformation and innovation. In some cases, that might change the migration focus or it may identify apps that could generate business savings to help invest in the business transformation and innovation.
Regardless of the apps that companies migrate or don’t migrate, center the focus on driving business innovation rather than migrating to the cloud because the competition is moving or because the CEO mandated a move. The focus on business innovation will drive greater success in the transformation to the cloud.
Sean Bryson is a high-tech executive and consultant focused on innovative technologies.