Data centers are getting very expensive to build and maintain. At the same time, though, data latency is hampering the advanced analytics and sensor-driven workflows that are the hallmarks of Big Data and the Internet of Things (IoT).
This is adding up to be a perfect storm for the centralized, on-premises data center and is forcing data professionals to rethink their views on the way data infrastructure should be optimized for an increasingly diverse set of user expectations.
This is why many organizations are turning to edge-based processing and analytics capabilities. By keeping the entire data stack on the edge, the twin problems of latency and complexity in the broader data environment can be addressed, because centralized resources are spared the overwhelming flood of data coming in from tens of thousands, and perhaps millions, of data points.https://o1.qnsr.com/log/p.gif?;n=203;c=204663295;s=11915;x=7936;f=201904081034270;u=j;z=TIMESTAMP;a=20410779;e=i
These all-in-one “micro data centers” are already evolving along a number of diverse paths to meet a range of specialized deployment options and use cases. Schneider Electric’s SmartBunker family is available in three versions: a traditional IT room solution, a module optimized for branch offices and a ruggedized field device. As well, the company has released the SmartShelter multi-rack model capable of functioning in multiple environments. The portfolio is due to be launched into global channels later this year, with the expectation that by 2018, it will play a central role in the roughly 40 percent of worldwide data traffic produced by IoT applications.
Meanwhile, a wave of start-ups is taking direct aim on the edge by completely re-imagining modular data infrastructure. Vapor IO recently emerged from stealth with a new “hyper-collapsed” design featuring a cylindrical configuration that packs power, cooling, compute, storage and networking within a 9-foot form factor. At the same time, the company is contributing its Open Data Center Runtime Environment (Open DCRE) to Facebook’s Open Compute Platform initiative as a way to track operational metrics across hyperscale environments. In this way, the company hopes to provide the foundation for the myriad processing entities out on the edge as well as the truly large-scale centralized facilities that provide regional services to what is expected to be a largely data center free enterprise industry.
The need to push resources out of the traditional data center is helping to fuel demand for cloud and colocation services, according to Enterprise Tech’s George Leopold. The advent of software-defined networking (SDN) is allowing entire data stacks to be elevated to the virtual plane, making it easier to integrate disparate resource sets into a working environment. At the same time, increased use of hybrid and all-Flash storage solutions are fueling the rise of real-time application performance that leverages the IoT to drive point-of-sale applications and other time-sensitive functions.
None of this is lost on the growing legions of cloud providers, of course. In fact, many are targeting edge services as a key growth area and are rapidly building out infrastructure to meet demand. Within two years, EdgeConnex has gone from two facilities to 20 and is expected to add another 10 by the end of 2015. These are not simply modular edge devices, says Data Center Knowledge’s Jason Verge, but facilities measuring up to 40,000 square feet. The idea is to provide a quick way for clients to extend infrastructure to where it is needed while providing a higher level of service than a standard content delivery network.
Regardless of whether data processing remains centralized or pushed to the edge, many of the traditional management rules apply. You still need broad visibility, systems monitoring and a means to remotely configure and reconfigure operating elements to adapt to changing conditions, and there must still be a way to coordinate the activities of these disparate entities to create an integrated data environment.
But increasingly, the edge is where the action is and enterprises that hope to remain relevant in the new data economy need to start placing their trust in the digital ecosystem that is emerging beyond the data center walls.
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, Carpathia and NetMagic.