Many articles and blogs about hybrid cloud computing are tempered with an abundance of caution. They turn back the clock, citing development/testing and rapid provisioning/de-provisioning as top reasons for using third-party cloud services. Heated debate persists about cloud security and compliance concerns, as if no progress had been made over the past two to three years. This is old stuff.
No reasonable person should throw caution to the wind when evaluating cloud service providers to augment internal private cloud infrastructures, thus creating an effective use case for their hybrid cloud. However, we can’t seem to get past the notion that it’s not just one big public cloud out there and hasn’t been for some time now.
Production workloads are getting done in the cloud today. Critical data with compliance requirements is being processed and stored there. The cloud still has all of the cost and computing benefits it had before, but now there are so many more reasons, options, models and capabilities to help drive business success. What any given customer requires in its hybrid cloud solution may not be available in the mega-public clouds (especially critical data storage and location), but it is definitely available.
A healthcare executive was quoted in a recent Forbes.com article, saying that HIPAA requirements prevent him from using a public cloud because regulators would not be satisfied. He was convinced, however, that it’s only a matter of time before the technology matures sufficiently to allow him to act. The thing is, HIPAA-compliant Infrastructure as a Service (IaaS) platforms and clouds are readily available in the market today. Instead of waiting for industry giants to catch up, he could be two to three years ahead of his contemporaries.
It’s a Hybrid World
The case for hybrid cloud infrastructures has been made. In just a few years, this cloud computing model will dominate. The internal or hosted private cloud, interacting and exchanging workloads seamlessly with a variety other service providers, will be the norm. It’s already beginning to happen.
There is no getting around the fact that some companies feel they must keep critical and sensitive data on premise. There is also no getting around the reality that legacy systems and applications continue to provide business- and mission-critical services and would be too disruptive and/or too expensive to re-platform for the external cloud. And, simply walking away from massive IT infrastructure and equipment investments is not practical. These are the companies that see hybrid cloud solutions striking an optimum balance between in-house security and the cost-effective flexibility of external service providers.
Hybrid solutions also provide IT management the optimum means for migrating workloads to external providers over time in accordance with their comfort and the maturation of the industry. This hybrid mode of deployment gives the enterprise the best of both worlds: managing the cost-capacity equation through the cloud, while retaining critical control within the enterprise.
What to Do Today
Hybrid solutions enable communication between on-premise services and the public cloud in order to allow the enterprise to retain critical control/critical data on-premise while enjoying the benefits of the cloud environment. Actually being on-premise isn’t necessary.
The private cloud component can be hosted by an IaaS and managed services provider and controlled by the customer. Aside from using the facilities and infrastructure of a third party, this model allows the user to have their private and public cloud services immediately adjacent, which mitigates distance-related latency and network bottlenecks.
However, without a geo-redundant private cloud configuration, organizations are vulnerable to catastrophic events. The customer could serve as the secondary site for disaster recovery and business continuity. Or, the cloud services provider could back up the data and maintain warm recovery at another data center a distance away.
Storage is an excellent use case for the hybrid cloud. Traditionally, all enterprise data was stored within the data center. With the exponential growth in data in all forms – records, texts, photos, audio and video – storing on-premise has become a costly proposition. The frequency of access for much of these data can be very low, but required duration could be very long. All cloud providers of substance today have layered storage options available, such as capacity storage and performance/high performance storage. Storage can remain connected as a service with the on-premise systems and ensure that low-frequency access data is pushed to capacity tiers, having tremendous potential to lower the total cost of storage management in enterprises of all sizes.
The now-you-need-it-now-you-don’t flexibility of cloud computing has given rise to cloud bursting. Cloud bursting refers to a situation when a spike in computing demand exceeds the safe bounds of internal capacity and workloads migrate to a different cloud environment to meet capacity demands. The spike can be anticipated, as in a month-end closing or seasonal traffic, or unexpected, driven by an event of some type. In both cases, adding capacity that will be underutilized most of the time is a waste of resources.
In a hybrid cloud, the steady state is handled by the fixed private cloud environment and the spike is absorbed by on-demand resources from a public cloud. For cloud bursting to be practical, organizations need secure, low-latency network connections between private and public clouds. As noted earlier, housing a private cloud in close physical proximity to a public cloud helps reduce latency issues.
Again, the outsourced cloud – depending on who the provider is – is not limited to these scenarios. We are at the point where full production IT environments operate in the cloud, as do workloads that demand audit-ready security and compliance compute and storage environments. The less that is required of the private cloud in the hybrid scenario, the greater the flexibility and efficiency and the more value to be extracted from I&O (infrastructure and operations) budgets.