6m read time
Cloud remains a top priority, with many companies sold on benefits ranging from cost savings to ease of deployment. Yet while cloud delivers these advantages and more, a variety of common misperceptions can trip up what it takes to migrate to cloud successfully.
One of the biggest misnomers is that cloud is easy—a simple “lift and shift” of core applications to the cloud, and organizations are off and running. The reality is much more complex: Yes, companies can flip the switch on Software-as-a-Service (SaaS) apps in no time. But in order to reap the full benefits of a cloud operating model, CIOs must design a well-orchestrated plan aligned to business objectives and pay close attention to where cloud fits within an enterprise architecture.
“Engineers that look at cloud deployments as the goal are missing a significant opportunity in driving agility,” says Issac Sacolick (@nyike), president of StarCIO and author of Driving Digital. “The reality is that selecting the appropriate cloud platform, architecture, and automation tools can have a long-lasting impact on the performance, reliability, security, and cost of managing applications and services.”
There is no one-size-fits-all cloud deployment model, and organizations need to think long and hard about which (public, private, hybrid, or multicloud) best maps to their needs while reducing time to value. “The economies of scale the cloud offers won’t be realized unless companies do some serious thinking and planning before jumping in,” adds Gene De Libero (@GeneDeLibero), chief strategy officer at GeekHive.
At the same time, there are barriers to an all-cloud approach–from having to integrate with legacy environments to network latency issues for particular applications. “A big misconception about the cloud is that all IT can migrate, which is simply not the case,” says Tony Flath (@TmanSpeaks), president of TmanSpeaks Ltd. “Hybrid design will reign for some time.”
Miscues over the true cost of cloud
Often times, planning deficiencies shine a light on another big cloud misnomer—that the model is inherently cheaper than traditional on-premises computing. In fact, unexpected or hidden costs can quickly mount up in the cloud, driven by bandwidth or storage requirements and by expensive new talent, according to Frank Cutitta (@fcutitta), CEO and founder of HealthTech Decisions Lab.
For example, many companies don’t account for the need for new skill training in areas like DevOps or information security, notes George Gerchow (@Georgegerchow), chief security officer at Sumo Logic. In addition, companies don’t adequately consider that applications must be properly architected to leverage cloud-native services.
Another common mistake is assuming a one-to-one conversion—that the number of cores and CPUs running on-premises are a direct correlation to the optimal cloud environment, which can quickly lead to expensive over-engineering, notes Jason James (@itlinchpin), CIO of Net Health.
“Cost engineering needs to be a continuous part of a strategy,” says IT Director Mike D. Kail (@mdkail). “The proverbial `lift and shift’ of an on-premises application will undoubtedly have adverse performance issues as well as a nasty monthly cloud bill.”
To avoid surprises, companies should ensure they don’t get locked into expensive contracts with providers—particularly those that might have painful penalties at the end. That’s especially true for companies looking to cut costs in response to economic challenges brought on by the pandemic, says Audrey Desisto (@AudreyDesisto), CEO of Digital Marketing Stream. “Workers may need to continue to work remotely and there could be penalties and costs to restart a project,” she says. “Make sure to read the fine print.”
Incomplete information on cloud security
Beyond confusion over cost, companies are also murky about cloud security, particularly their role in the shared responsibility model, notes Will Kelly (@willkelly), technical marketing manager at Anchore. Conventional wisdom is that cloud vendors are completely responsible for security, but depending on the vendor and type of cloud service used, the enterprise maintains ownership of certain security aspects.
Without full knowledge of what’s required, companies may have to come up with quick-turn cloud security plan that can take a sizeable bite out of budgets, cautions Jones Baraza (@JonesBaraza), director at CyberSpace Kenya.
“It’s not as simple as my data and applications are in the cloud now, so everything is secure,” adds Debra Baker (@deb_infosec), CISSP CCSP and a cybersecurity evangelist. “You must understand what cloud service is being used and the shared responsibility to secure the applications and data being stored in the cloud.”
Proactively addressing data architecture issues is another important element of streamlining cloud migration and keeping a lid on costs. Most companies get caught up in trying to figure out which applications to move to the cloud, in what order and how. But while applications generally remain static, data volumes do not. “During migration to the cloud, most organizations think they have an application problem … but once they’re in the cloud, they find they actually have a data problem,” says Bill Mew (@BillMew), founder and CEO of Crisis Team. “Without a clear data architecture, it’s impossible to see where data is duplicated or to differentiate between data that needs to be accessed regularly, data that can be archived, and data that can be discarded.”
Creating a cloud culture
Part and parcel of the cloud journey is creating the right culture, including reorienting IT to stop thinking of cloud as outsourcing and consider going cloud-first to drive speed, agility, and the ability to easily scale, says David Chou (@dchou1107), a global health care CIO.
At face value, cloud may be easier to deploy, but that very fact means organizations need to be more rigorous about tools and processes when managing deployments, adds Mark Thiele (@mthiele10), CEO of Edgevana.
IT operations also need to transition away from long-term big projects to planning, building, and iterating smaller initiatives, including adoption of DevOps practices that tie together automation and project management. “Practices like `site reliability engineering’ enter the lexicon of business to instrument applications and quickly report and respond with converged engineering, security, and infrastructure expertise,” says Wayne Anderson (@DigitalSecArch), security and compliance architect at Microsoft 365 Center of Excellence.
At the end of the day cloud is not a substitute for having a real solutions strategy. “You can’t just throw everything over the wall and assume I’m in the cloud so everything will work great,” says Jack Gold (@jckgld), principal analyst and founder at J. Gold Associates LLC. “For cloud implementations to be successful, it takes work and planning just like on-premises solutions.”
To learn how Lenovo can help you migrate to cloud successfully, click here.
Organizations today are under unprecedented pressure to adapt. Intel has delivered five generations of custom silicon built for cloud scale, along with co-engineering with partners and relationships with top cloud providers. Today’s top clouds are powered by Intel® Xeon® Scalable Processors.