A few years ago, it appeared that the public cloud was going to take over the enterprise. That didn’t happen, but neither did the enterprise fall completely to the private cloud or the hybrid cloud. In fact, it seems that the data center industry is still feeling its way through the conversion from hardware-centric to software-centric infrastructure, leaving open the very real possibility that no single architecture will dominate the future.
According to IDC, revenue from public cloud services topped $63 billion in the first half of the year, a 28.6 percent gain over the same period in 2016. More importantly, says analyst Frank Gens, public infrastructure has become the launch pad for a number of data initiatives, such as artificial intelligence, blockchain and quantum computing, that are expected to play major roles in enterprise operating models going forward. And perhaps not surprisingly, the highest growth for public cloud comes from the Asia Pacific region, which is probably due to the relatively small installed base of traditional IT infrastructure.
But this expansion masks the reality of the public cloud that many enterprises come to realize only through experience with production environments. As Densify’s Yama Habibzai noted on Information Age recently, misconceptions surrounding costs, ease of management and flexibility abound, particularly when it comes to linking internal workflows to proprietary clouds like AWS. Part of this can be attributed to the rush to get on the cloud, which led to many workflows migrating in a haphazard fashion before proper management and governance tools were established, but it is also due to the fact that renting infrastructure is not always cheaper than buying, particularly when operations begin to scale.
Another issue that many public cloud users fail to appreciate at the initial deployment is the need for a strong exit strategy, says Forbes’ Avishai Sharlin. As competitive pressures pushed the enterprise onto the cloud, little thought was given to optimizing the right cloud for the right workload, and without taking care to ensure that their new cloud apps had the proper APIs and industry-standard software configurations, many organizations were locked into the same ill-fitting infrastructure silos they were hoping to get away from. It is heartening to see many users adopting multi-cloud and service-broker strategies, but it will take some time before this change runs its course.
Of course, this is the thinking that is driving the current wave of hybrid cloud adoption, which, according to IDG, will dominate the future. But even many of those who subscribe to this notion admit that hybrids are no panacea for all enterprise workloads. In the first place, hybrid infrastructure requires broad conversion of legacy data systems to virtual, cloud footings, which is neither easy nor cheap. Secondly, the level of automation needed to seamlessly integrate internal and external resources is only just starting to take shape, so hybrids will remain an operational challenge for a while longer. And thirdly, even hybrids require specialized application coding to accommodate the expanded resource options, which means wholesale conversion of the full enterprise suite will be a slow and tedious process.
None of these challenges are insurmountable, of course, but it is safe to say that the future of data infrastructure is not nearly on the clear and steady path that either public or private cloud proponents imagine. About the only thing we can say for sure is that the data environment will become increasingly virtualized and distributed, first to the cloud and then to the edge, and that each enterprise will be responsible for crafting their own software-defined architectures to handle their unique needs.
The days of simply ordering off-the-shelf hardware or generic virtual resources are coming to a close.
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.