Now that we’ve set the framework for hybrid cloud, it’s important to outline the advantages and disadvantages of this approach. If you recall from the earlier post on The Good, The Bad, and The Hybrid, Stephen Hawking famously said, “Without imperfection, neither you nor I would exist.” Conversely, hybrid cloud, though it marries the benefits of private and public, also carries many caveats and is not a perfect solution, especially if execution misses the mark.
First, optimistically, hybrid cloud unlocks immense opportunity for business by providing scale, capabilities, and feature sets to adapt and change direction quickly while building for future demand – the fundamentals of technological agility. The ability to turn on a dime and implement new digital experiences increases speed to market and shortens time to value, allowing business to keep up with the unique and challenging demands of the market.
Hybrid cloud also provides reliabilities in service delivery that are difficult, if not impossible, to realize and achieve in a private cloud model. It allows businesses to leverage geographically dispersed platforms to deliver services closer to technology consumers while overcoming latency and continuity concerns.
Further, hybrid cloud creates a “buffet of technology services” that may be consumed when and wherever required, allowing for efficient scale up, scale out, and even scale down, which transforms into significant cost savings on both capital and operational spend.
Now, donning the pessimist hat, hybrid cloud brings about many roadblocks to successful adoption. At the forefront, skillsets have been found to be the biggest obstacle towards hybrid cloud adoption. The demand for talented IT architects, engineers, and analysts with experience in public cloud platforms does not meet supply, and the onus is placed on businesses to either develop the appropriate skillsets internally, and/or engage with third party organizations, such as Third Octet, to assist partially or entirely on developing the hybrid cloud strategy through architecture to implementation.
Hybrid Cloud, at the end of the day, is simply the unison of private and public cloud and still carries all the cost concerns with each model – the heavy capital hits of private cloud infrastructure spend and the operationally intense consumption model of public cloud. Businesses must still understand where all cost implications exist within the hybrid model and thoroughly monitor consumption to prevent any runaway costs.
In developing your hybrid cloud strategy, IT must also understand the integration approach and how private and public services will be linked to provide a common delivery strategy. Hybrid service delivery requires a symbiotic relationship between applications and data and, during integration, it is imperative to maintain these relationships without the impact of latency or isolation. As well, integration of services within public cloud could be cost prohibitive (i.e. as with significantly sized workloads); therefore, if a “cloud first” mentality exists, know that not all workloads could or should run in public cloud.
As businesses have traditionally built private cloud long before any consideration of public or hybrid cloud, one will face platform duality in a hybrid model where the platform built to support private cloud (i.e. hypervisor software) will more than likely not match the platform powering public cloud. As a result, if businesses are looking to leverage a hybrid model to support portability and compatibility of workloads across the hybrid platform, it may be necessary to account for transformational requirements when any such workload is moved between cloud platforms.
And, of course, extending a private cloud to a public cloud and defining the hybrid fabric requires new methods of connectivity and dependency on such. In a pure private cloud model, dependencies on external connectivity are not as great as services exist largely in a private interconnected environment; however, when services extend beyond the private cloud model, the requirement to ensure adequate performance and maintain near continuous availability translates into a significant requirement on external connectivity to public cloud services. Unfortunately, this critical requirement is often overlooked and represents a failure in execution of a hybrid cloud model.
Take, for example, a simple migration to Office 365. Previously, e-mail connectivity to and performance of a private cloud Microsoft Exchange was outstanding – the supporting infrastructure existed within a high-speed localized networking infrastructure. As the organization moved towards Microsoft Office 365, however, performance of e-mail connectivity degraded significantly as the existing external connectivity was not appropriately sized to now support the new demand driven by public cloud services.
Lastly, hybrid cloud not only extends one’s private cloud infrastructure, it also extends existing IT teams and creates dependencies on third party organizations. What once was a straightforward task to conduct internally, such as network troubleshooting, now becomes a potentially strenuous exercise where businesses depend on the service teams affiliated with the public cloud provider.
These are just a sampling of the advantages and disadvantages of hybrid cloud and, erroring on the side of caution, it’s important to thoroughly understand the disadvantages over the advantages as execution is usually the main obstacle towards successful hybrid cloud adoption. If businesses are aware of the challenges they’ll face on their cloud journey, such as scarce skillsets, businesses will be able to plan, understand the risks, and close the gaps to ensure successful execution immediately and over the life of hybrid cloud.
How does one get started with hybrid cloud? Tune in next week.