Edging toward remote management of edge data centers

Categories Edge Computing News  |  Guest Posts
Edging toward remote management of edge data centers

Guest post by Michael C. Skurla, Chief Technology Officer of Radix IoT

Data centers are expanding beyond large-format facilities and moving to the edge—and for very good reasons. A distributed architecture helps localize computing power closer to customers, decreasing latency, and expediting data transfer. Edge data centers also promise security and privacy that is more attuned to the needs of the many new data sources that are being connected to cloud ecosystems upstream. While edge data centers offer the ability to help corral these new data sources, managing the proliferation of data centers will require a new approach to facilities management.

One of the major drivers of the market demand for edge data centers is the increased adoption of compute-intensive IoT and AI applications. Latency has been a key bottleneck for large factor data centers and has been a critical concern as more and more devices onboard into ecosystems than ever before to drive IoT applications and analytics. For these applications, however, latency is simply not an option.

Edge data centers meet IoT application demands with localized computing power and drastically lower data processing latency. By 2025, Gartner predicts 75% of enterprise-generated data will be created and processed outside a traditional centralized data center.

“Organizations that have embarked on a digital business journey have realized that a more decentralized approach is required to address digital business infrastructure requirements,” Santhosh Rao, senior research director at Gartner has stated. While decentralization of applications solves one set of issues, they raise another: how to manage decentralized data centers. Understanding what an edge data center is will help enterprises find a way to answer the facilities management question.

What is an Edge Data Center?

An edge data center is often misunderstood as a replacement for cloud or larger data center infrastructure. Nothing is further from the truth. Edge data centers can be identified typically by the following characteristics:

– They usually measure in feet, not acres, typically having 8-10 racks, (sometimes even down to 1 or half a rack) and having a power profile of 100KW at most.

– They are highly localized to the region they serve. Typically, service providers are not putting an edge data center in Toronto to service Chicago. In fact, this can even go down to smaller units. An edge data center could easily be dedicated to a few city blocks.

– Typically built out in interconnected networks of edge data centers to cover geography and work in tandem with a larger infrastructure, often in colocation or hyperscale data centers.

– Mission-critical facilities—edge data centers have an exceptionally low tolerance for failure or outages as they are serving infrastructure that is deemed essential, or near essential for the public or private enterprise applications.

– Free of on-site staff, and completely remotely managed and monitored. They are set up across the globe, often in hard-to-reach areas. While they include all the standard building technologies, edge data centers include multiple software solutions from different vendors to make them operate in this autonomous fashion. These software solutions are typically networked together using next-generation IoT platforms to make remote rollout and management possible, whether it’s one site or thousands.

Using IoT platforms to manage the edge

Edge data centers use IoT platforms to seamlessly maintain, monitor and manage this critical infrastructure from a central location. IoT platforms, compared to older BMS or SCADA style solutions, offer operators the ability to capture, organize, and analyze data from all the technologies and subsystems used to keep the facility alive without local human intervention. The platform turns this data into useful, actionable, outcome-based analytics. From improved issue response to predictive maintenance, security, uptime, and control of operating expenses, operators can use the unified dashboard of a single pane of glass to seamlessly and efficiently monitor and manage varied equipment across geographically distributed locations–all done remotely.

Armed with data-driven IoT platforms, operators can not only monitor and manage their critical infrastructure across the globe but troubleshoot and maintain business continuity—eliminating costly downtime. How costly? The average cost of network downtime is around $5,600 per minute and nearly $300,000 per hour according to Gartner. When compounded with a portfolio of remote locations and the associated cost of truck rolls, monitoring is vital.

With downtime not an option, IoT platforms are the most cost-effective solution to disaster-proof critical assets. Before major problems turn into irreversible disasters and exorbitant costs, operators get alert notifications of risks. IoT platforms can also be modified and turned on the fly to adapt to changing environments or equipment, all remotely. This ability to adapt and divert risks promptly, while also allowing remote triage, drastically lowers operational costs—more importantly, it allows for uninterrupted uptime.

IoT platforms Seamlessly Integrate With Current Systems

Open source by nature, IoT platforms are vendor-agnostic solutions allowing varied devices from different vendors to flawlessly work together without vendor lock. While IoT platforms don’t replace a BMS or other purpose-built solutions for individual trades, they allow consolidated management of all the systems together in synergy.

IoT platforms consolidate all the data into a single, organized data lake, allowing higher flexibility of management, remote access, monitoring, and control that users can easily change on the fly. This consolidated dataset can also seamlessly connect to external micro-service analytics, software tools, or BI engines to provide actionable analytics offered by no single trade’s data silo. And it easily adapts to the specific monitoring or control requirements of critical facilities.

Edge data centers are here to stay; indeed, they will only multiply. And clearly, IoT platforms are the most cost-effective solutions to expedite rapid provisioning of newly deployed facilities on a global basis. Consider IoT platforms to be the unifying layer over existing infrastructure—without requiring rewiring, replacement, or ripping out existing systems. As an abstraction of data connectivity easily added, IoT platforms empower facility operators by providing full access to actionable data, cost-effective monitoring and control of their facilities—and the ability to do it all remotely in the comfort and safety of social distancing.

About the author

Michael C. Skurla is the Chief Technology Officer of Radix IoT — offering limitless monitoring and management rooted in intelligence — and sets cutting-edge product strategy for the company’s IoT platform. He has over two decades of experience in control automation and IoT product design with Fortune 500 companies, focusing on the intersection of software and hardware that emphasizes data aggregation and analytics for mission-critical industries. Skurla is a contributing member of ASHRAE, USGBC, and IES Education.

DISCLAIMER: Guest posts are submitted content. The views expressed in this blog are that of the author and don’t necessarily reflect the views of Edge Industry Review (EdgeIR.com).

Article Topics

 |   |   |   | 


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Featured Edge Computing Company

Edge Ecosystem Videos

Machine learning at the Edge


Latest News