It may have taken a little while, but a critical mass of enterprises has now developed private cloud architectures at sufficient scale to start thinking about tying them to public resources to create the so-called hybrid cloud.https://o1.qnsr.com/log/p.gif?;n=203;c=204663295;s=11915;x=7936;f=201904081034270;u=j;z=TIMESTAMP;a=20410779;e=iAnd in traditional IT fashion, it turns out that the reality of this moment is quite a bit different from the expectation. Not only are hybrids more complicated than originally thought, there is growing suspicion that the rationales for creating them in the first place are not all they were cracked up to be.
In the first place, it is becoming clear that cloud architectures – whether public, private or hybrid – will not follow the standard generic infrastructure model of legacy infrastructure. Rather, hybrids will be built from the ground up with specific use cases in mind, which means they will differ in both form and function depending on the applications and processes they are to support.
According to ZDnet’s Manek Dubash, the enterprise will have to choose very carefully among a multitude of cloud implementations depending upon key application requirements like security, residency and performance. At the same time, the overall cloud infrastructure must remain fluid enough to accommodate new services and applications as they are introduced, as well as existing ones as they evolve over time. So not only will it be challenging to implement the kind of hybrid cloud you need right now, it will only become more so as the environment scales and develops.
At the moment, at least, there appear to be five general flavors of hybrid clouds, says InfoWorld’s Eric Knorr, but there is ample reason to believe that more will arise. The most common is a self-service duplication cloud, which places a carbon copy of on-premises architectures on public resources for data bursting and other application needs. At the other end of the spectrum is the on-premises public cloud, championed largely by Microsoft, that bundles a complete cloud environment on local infrastructure, essentially allowing public services like Azure to occupy part of the enterprise data center. In between are hybrid application architectures, virtual machines on-demand, and managed multicloud solutions, each of which provides various levels of scale, resource distribution and management integration.
To hear tech consultant Keith Townsend describe it, however, none of these approaches has a compelling use case because the operating environments of local and cloud-based systems simply do not lend themselves to integrated application support. The industry has already accepted the fact that applications designed for traditional data architectures do not work well in the cloud without a lot of recoding, and vice versa, so how are they expected to transcend these two worlds seamlessly and in support of common goals? In all likelihood, public and private environments will run parallel to one another, perhaps with some level of data exchange but ultimately supporting separate applications running unique workloads for highly targeted outcomes.
Unless, of course, a new set of applications were to emerge that is uniquely suited to hybrids. For Michael Davison, financial industry principal at IT services firm ATOS, that could very well be the advanced analytics that organizations will need to run Big Data and IoT workloads. Analytics as a Service, in fact, is already fairly advanced in financial services, and placing it on a hybrid model will afford organizations with the scale and control needed for their highly regulated business models. Ultimately, the hybrid cloud may be the only way to enable the integration of IT and business processes, which will be crucial in an increasingly service-driven economy.
Clearly, it is too soon to close the book on the hybrid cloud, but neither is it a small dunk. The technology certainly exists, for the most part anyway, but the creative forces within the enterprise have yet to leverage it in a truly meaningful way.
But this is not unusual, since virtually every major advance of the past needed a certain gestation period before its transformative capabilities could be fully realized. This may or may not happen with the hybrid cloud, but even if it doesn’t, it is fair to say that whatever emerges instead will find success based on the solutions it provides, not the technology it employs.
Arthur Cole writes about infrastructure for IT Business Edge. Cole has been covering the high-tech media and computing industries for more than 20 years, having served as editor of TV Technology, Video Technology News, Internet News and Multimedia Weekly. His contributions have appeared in Communications Today and Enterprise Networking Planet and as web content for numerous high-tech clients like TwinStrata and Carpathia. Follow Art on Twitter @acole602.