The rise of cloud computing, like any evolutionary IT trend, has had its fair share of hype – and the attendant noise that leads to confusion and myth-making.
Hybrid cloud is no exception: The hybrid approach has spawned a variety of misconceptions. We asked several cloud experts to share their perspectives on some of the key misunderstandings about hybrid cloud that persist. Here’s what they had to say:
Misconception 1: Hybrid cloud means sacrificing governance and security.
It has been a longstanding – and sometimes overwrought – fear in enterprise IT: Cloud computing means a loss of control that is ultimately detrimental to governance, security, and related areas. But for many organizations, that concern is misguided, outdated, or both. In fact, the hybrid cloud approach may offer the best path to unlocking the potential of cloud while retaining close control over certain types of data or workloads.
[ What is multi-cloud and what it do for you? See our related article, Multi-cloud vs. hybrid cloud: What's the difference? ]
“This myth certainly had legs in the early days of cloud computing, when offerings were less customizable and IT departments had little control over management or visibility,” says Kong Yang, Head Geek at SolarWinds. “However, while migrating applications and workloads to the cloud does relinquish some control, particularly day-to-day maintenance, organizations today have more freedom to construct the solutions they need as cloud providers continue to reduce the barrier to consume cloud services.”
Yang certainly recommends that you do your due diligence on providers and platforms, and subsequently keep current with new features, functionality, and so forth. But cloud has significantly matured since the earlier days where outages and downtime were more common; don’t let fear dictate your IT strategy, especially when it might be misplaced.
Misconception 2: Hybrid cloud won’t deliver the same level of performance and availability as exclusively on-premises infrastructure.
Yang also points out that, as recently as several years ago, there was some truth to the idea that cloud environments were generally less reliable than on-premises systems. But as cloud has matured and the tools and services for effectively managing cloud systems have likewise grown more abundant and much stronger, this "truth" is becoming fiction.
“Today’s cloud provider SLAs, combined with the simplicity of setting replicas, standby systems, and the durability of data stored in the cloud, often outperform what IT departments can deliver with an average on-premises system,” Yang says.
Misconception 3: Significant on-premises infrastructure investments preclude cloud adoption.
In a recent article on hybrid cloud trends, we described how some IT leaders and teams are flipping a well-worn script by treating legacy infrastructure as a catalyst of cloud projects rather than an obstacle. In a similar vein, there’s a notion that extensive (and often expensive) investments in traditional infrastructure are a barrier to cloud adoption, according to PITSS director of consulting Gavin Woods.
“One common misconception is that [companies that] have invested in on-premises technology stacks for their systems and [their] support and maintenance teams think they can't use cloud-based technology,” Woods says. In reality, Woods says that with the exception of especially monolithic legacy systems, cloud projects don’t require scrapping existing investments or doing cumbersome modernization overhauls. It’s a matter of choosing the right software tools and implementation standards, as well as the correct cloud architecture for your organization, he adds. Moreover, the hybrid approach may actually be the most attainable strategy for companies with considerable on-premises investments.
“For a lot of organizations [that] need to migrate legacy/existing systems running on-premises to the cloud, a hybrid cloud architecture is a feasible next step, instead of going big-bang in migrating every piece of your technology stack to cloud-based technology,” Woods says.
Your company’s move to a hybrid cloud model, or update to your hybrid cloud model, also begs for a thorough examination of the application portfolio, as E.G. Nadhan, Red Hat’s chief technology strategist (Central Region) recently noted. (See our related article, “Hybrid cloud projects: 8 questions to ask first”) Says Nadhan: “You need to give more attention to enabling applications for significant business functions during your move to the hybrid cloud model. Now’s the time to identify the applications that really matter. You also want to identify redundant functionality across applications and rationalize them down to a more manageable number. Some applications may have to be decommissioned, while others could merit a more state-of-the-art, user-friendly interface.”
“This is also a good time to revisit the original business requirements for existing applications and revise them to cater to the needs of newer generations of users,” Nadhan adds. Some, of course, may be best left alone.
Misconception 4: All cloud environments have a lot in common.
The proliferation of cloud platforms and services can make all of your options seem to blur together: They’re probably all roughly the same thing, right? No. In reality, the hybrid cloud approach itself is a pragmatic acknowledgement that not all clouds are the same, and especially not when comparing private and public options, just as on-premises infrastructure also comes with particular considerations.
“Each of the clouds in market today are different technically, architecturally, commercially, and in many other ways,” says Jeff Budge, VP of advisory consulting & product management at OneNeck IT Solutions. “While appropriate workload placement is highly important in the hybrid world, the differences between the clouds go deeper than the high-level perspective of workload suitability to a particular cloud.”
Budge recommends doing your homework on several other considerations for your hybrid portfolio, including:
- What the cloud platform will require you to architect and procure to truly provide a highly available application: “At OneNeck, we have seen vast differences in the complexity, cost, and supportability of the same workload when architected to the same level of availability on different cloud platforms,” Budge says.
- Networking is still relatively young in the cloud world: “Many cloud providers created many attractive options for compute, storage, and security in their service catalog,” Budge says. “However, we have often found that the networking options are not as robust or flexible.”
- Pricing models continue to evolve from the basic to more complex and compartmentalized models: “Perhaps someone will be bold enough to create a less a la carte pricing model and return us all to a more predictable state,” Budge says.
- Terms and conditions are not as similar from cloud provider-to-cloud provider as you might think: “Be aware of language that allows the cloud provider to make changes to the underlying platform, especially under short notice,” Budge advises.
