Quest For The Hybrid Grail: Getting To A Frictionless Cloud
August 16, 2021 Satya Sharma
Grail quests have always been a motif in literature and entertainment, and they most always come about because of an existential threat to the heroes seeking the mystical relic. In many ways, grail quests are also equally common in the business world. Businesses have found that a highway to accelerate their quest is embracing hybrid cloud, which recent studies have shown to offer 2.5 times greater value than a public cloud only deployment. But not all hybrid cloud platforms are created equal.
In an IDC post titled The Road to Hybrid Multicloud, the market researcher explains that “the Holy Grail in cloud computing is a frictionless, hybrid multicloud approach that provides consistent experience and unified management across multiple public clouds, private clouds, and even traditional infrastructure.” While many organizations are still striving for this goal, let’s unpack what it means to be “frictionless” and the pitfalls of not adopting a frictionless hybrid cloud.
The Friction Of Applications And Management
When we look at the market and talk to our clients, we find that many of their most important, mission-critical applications still reside on-premises in the data center, while they are embarking on public cloud usage for some of their use cases. Often, these applications are developed in-house, “home grown” or highly customized ISV products to meet a business’s unique needs, and when they consider moving these applications, the costs of refactoring and porting them into a public cloud outweighs the benefits. As the organization continues to expand, this can lead to cascading effect as the siloed but critical on-premises applications become more and more isolated, and IT administrators spend more and more time managing applications across different IT architectures.
In a frictionless world, a hybrid cloud approach can easily accommodate architectural consistency. This means that users no longer must worry about silos because their existing on-premises architecture doesn’t match that of their cloud provider, nor does it mean that they can’t select the right architecture to get the most out of a specific workload. A hybrid cloud platform with architectural consistency in both on-premises and in the cloud gives them the best of all worlds and allows them to be one step closer to frictionless.
As a bonus, incorporating a hybrid cloud strategy has the potential to speed up DevOps, by allowing your team to instantly spin up a containerized environment for a new application, or use it to deploy a new modernized or cloud-native application and connect it to your production workloads for a very low cost. This means you can experiment with more new applications and seamlessly integrate them into production workloads.
The Friction Of Scaling
One thing not covered in IDC’s definition of a frictionless is scaling, and as we saw in the dramatic and sudden shift in standard business practices brought on by the pandemic, frictionless scaling of a hybrid cloud is essential for businesses. Now you may be thinking, on a public cloud scaling is as easy as selecting and swiping a credit card, and while this may sound good on paper, in practice it’s hardly ever that smooth. That’s because many of an organization’s mission-critical applications are attached to large amounts of data, so if you experience a surge in demand for those applications, where the data resides and how it is accessed needs to be considered. That’s why we’ve expanded our definition of a frictionless hybrid cloud to include frictionless scaling, regardless of where the resources you’re trying to scale are located.
A truly frictionless hybrid cloud architecture extends the cloud-like scaling capabilities of the public cloud across the entire hybrid infrastructure, with the ability to scale up and down for on-premises systems with a simple button click, as well as the ability to scale your cloud infrastructure within the same dashboard. Within this dashboard, frictionless hybrid cloud users will acquire and spend a single currency that can instantly unlock capacity across the hybrid environment, both on-premises and off and dynamically move that capacity across resources, so that the workload needed during a surge can access additional resources to manage, and then scale back down when no longer needed. With many common applications basing their license fee on core counts, the ability to dynamically activate and deactivate cores can also improve cost savings for organizations as well.
Putting It All Together
So, what would this mean when it’s put into practice? While this is just our vision, let’s picture a scenario in the not-too-distant future when King Arthur Industries is poised to release their brand-new microwave and dishwasher-safe grail, perfect for your (k)nightly beverage of choice, just in time for the holiday season. King Arthur Industries’ CIO Lance A. Lot knows to expect a drastic surge in demand both from the annual holiday shopping spree and the new product launch, and he wants to be prepared.
Knowing that his on-premises production environment has enough capacity to meet King Arthur’s needs eleven out of 12 months of the year, he doesn’t have the budget nor the time to go through the procurement process of expanding his on-premises server environment, but he has unused cores that he can use on a pay-per-use basis, and some pre-production and development environments that he can move to cloud. First, he quickly scales up his on-premises with pay-per-use capacity. He still needs more capacity, so thanks to the architectural consistency between his on-premises and cloud infrastructure, he seamlessly moves his pre-production and development environments to cloud-based resources, freeing up even more capacity while also allowing this pre-production workloads to continue running without interruption.
The market for grails booms, and King Arthur Industries easily handles the surge for their online marketplace and can meet every customer order on-time. As the surge dies down, Lance A. Lot brings down his infrastructure costs with pay-per-use model and brings the workloads he moved temporarily back on-premises and scales down his cloud presence. But he has written a successful game plan for the next surge, whether it be a predicted one like holiday shopping season, or an unexpected one where he needs to be able to move quickly to keep King Arthur Industries up and running.
Go Frictionless Or Go Home
When it comes to clouds, there is only one model that has the potential to deliver on the promise of cloud computing, and that’s hybrid. Of hybrid cloud models, only frictionless ones can deliver the holy grail to businesses. With the ability to manage applications across diverse resources and platforms, better enable DevOps to innovate new ways for the business to succeed and help drive efficiency and cost savings with dynamic scaling across the entire IT environment, whether on-premises or off. That’s the vision that we at IBM have for a frictionless hybrid cloud platform, and we’re hard at work to make it a reality for our customers.
Satya Sharma is an IBM Fellow and chief technology officer of Power Systems at IBM.