For a while, it seemed like the cloud was the end-all and be-all of enterprise infrastructure. With its unlimited resources, broad flexibility and low cost, how could technology get any better?
Of course, that’s what they said about the internet, the server, the PC and mainframe.
The fact is that with the emergence of the IoT and the distribution of data processing and storage to the edge, we are already laying the groundwork for the post-cloud era. The cloud will continue to exist, even thrive, just as the mainframe does today, but it will no longer be the primary driver of data productivity.
But it’s important to understand what constitutes a post-cloud development in terms of infrastructure vs. what is happening today on an operational level. China’s BaishanCloud, for instance, recently introduced a “cloud-linkage service,” which is basically an API interface that allows organizations to manage many clouds as one. Although this is necessary in a diverse cloud ecosystem, it’s a bit of a stretch to say, as VP Eddie Zhao does, that it targets a post-cloud environment.
Closer to the mark are the multitude of developments taking place on the edge, says venture capitalist Peter Levine. In an interview with Business Insider, he notes that even the cloud cannot cope with the reams of data coming in from literally trillions of connected devices, so by necessity the bulk of processing and storage will have to take place not just at the current edge but on the devices themselves. Cisco and others are already calling this “fog computing” but on a fundamental level it is the creation of yet another tier of infrastructure designed to support the entirely new data sets and application requirements that a device-driven ecosystem generates.
What will these device-level computers look like? At the University of Michigan, researchers are perfecting what they call “micromotes,” millimeter-sized computers that can run their own analytics and support their own applications – no cloud or edge necessary. The researchers unveiled the latest generation of micromotes at the IEEE’s International Solid-State Circuits Conference in San Francisco last month, demonstrating their ability to analyze sounds and temperature using only a few nanowatts of power. Other versions have been shown to maintain their own wireless connectivity up to 20 meters, which could turn, say, a smart automobile into a self-contained data ecosystem.
Such a system requires storage, of course, and housing it directly on the devices themselves will drive up cost and complexity. But a company called Open.IO just came out with a new architecture that it says addresses the data protection, resiliency and availability needs of the IoT. The system is built around a ServerLess Storage (SLS) appliance that features Linux-based “nano-nodes” that house a pair of ARM CPUs, RAM and Flash memory and dual Ethernet ports that connect to 40 GbE switches. The idea is to reduce the storage failure domain to a single node using a set of advanced algorithms that enable real-time cluster availability and other features that even object storage solutions cannot match. In this way, failover is managed in a matter of seconds and devices across a distributed architecture have ready access to data as if it were housed locally.
A post-cloud world marks the continuation of a decades long transition away from centralized data infrastructure to distributed architectures that will soon inhabit virtually everything around us, and perhaps inside us as well. While this may help share the data load, it will inevitably lead to a host of challenges surrounding control, management, security and connectivity.
At heart, the nature of computing will remain the same: Gather data, give it meaning, and then share it with others.
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.
