• United States

Stay smart as you approach the edge

Mar 23, 20184 mins
Data CenterInternet of ThingsNetworking

As computing moves ever nearer to the edge, delivering faster, more efficient services, the need for visibility across these new-look, complex networks become abundantly clear if businesses are to make the most of their new investment.

Fundamental to harnessing the full potential of the Internet of Things (IoT) is the need for decisions to be made in real time, and it’s in addressing this that discussions have turned to the subject of edge computing over recent years.

Before the data generated by myriad of connected IoT devices is sent to the centralized cloud, edge computing sees it stored and processed locally, in distributed micro-clouds at the edge of the network, closer to where the devices are placed, and the data produced. Doing so cuts down on the need for data traffic to be back-hauled to and from a remote data center, thus making it ideal for supporting the real time data delivery required by the IoT.

This is further optimized by the deployment of machine learning, effectively making edge compute devices “smarter”, empowering each one to train itself using the particular dataset it processes to conduct more effective analysis to glean actionable intelligence at the edge.

Edge computing also lends itself to improving the efficiency and efficacy of data-driven cloud-based enterprise applications and addresses several issues faced by businesses today. With data first processed by multiple interconnected micro-data centers close to the edge, before being sent to the cloud for additional processing on a global level, IT teams are afforded more application scalability and enjoying perhaps the greatest benefit of edge computing – significantly reduced latency. 

As businesses recognize the efficiency and benefits of moving compute processing power, storage and networking to the edge, the technology is rapidly gaining momentum. Qualcomm, for example, recently launched a suite of edge computing services, just a week after Intel announced its own entrance into the space.

Need for assurance

With edge computing set to underpin a growing number of applications affecting our lives and businesses, it’s crucial to have assurance that the connectivity of the network on which these applications run remains ubiquitous, consistent and responsive. But as with any burgeoning technology, there will always be a new set of challenges to face and dealing with these will require complete visibility across the entire lifecycle of these applications.

Unless traffic data is continuously processed at the edge, normalized and correlated in the context of service delivery, operations and business performance, it would be challenging to gain a real-time insight on application and service assurance.

Unfortunately, this situation is only set to become more complex once edge computing is applied, at which point traditional network monitoring and assurance tools may no longer be enough to provide companies with the window they need on this new environment.

The only way to gain the insight required for full and effective management of an organization’s network and the applications that run on it, therefore, is to employ a smart data solution which will maintain visibility throughout every aspect of the application’s lifecycle.  

Valuable intelligence

Smart data is real-time, scalable metadata imbued with context and variety of other attributes derived from deep analysis of network traffic, that offers unlimited scale, across all aspects of the network, both physical and virtual. Once a business can access and analyze this data in real-time, it will be able to gain valuable new insights into how the applications on its network behave, how they interact with the network, and the type of traffic patterns they produce. Ultimately, with access to smart data, businesses will be in a better position to make more informed decisions about how to optimize their applications and networks, where to allocate capacity, and how to boost performance to improve operational efficiencies.

Furthermore, the visibility enabled by smart data can also enable businesses to identify anomalies within the network; significant changes that could indicate issues such as threat agents and performance issues, which could potentially affect the running of hundreds or thousands of connected devices and applications. Actionable intelligence of this type is worth its weight in gold to an enterprise who relies on the network to support its IoT and business application deployments.

Potential for success

It may still be in its early stages, but the benefits of capacity, low-latency and scalability that edge computing offers represent great potential for the success of both the ever-expanding IoT and of businesses running much of their workload in the micro-clouds / micro-data centers.

As computing moves ever nearer to the edge, delivering faster, more efficient services, the need for visibility across these new-look, complex networks become abundantly clear if businesses are to make the most of their new investment.

Only by employing a smart data solution will they have the ability they need to monitor the sheer breadth and depth of their application ecosystem, safe in the knowledge that everything is assured.


As area vice president of strategic marketing at NETSCOUT, Michael Segal is responsible for market research, enterprise solutions marketing, analyst relations, customer advocacy, analyst relations, advertising, and social marketing.

Michael’s product management experience spans across ten years at Cisco Systems, where he managed all aspects of product line life cycles for several successful product lines. Michael's technical areas of expertise include SaaS/cloud, virtualization, mobile IP, security, IP networking, Wi-Fi/wireless, VoIP, and remote access. Michael holds patents in areas of networking and wireless mobility.

The opinions expressed in this blog are those of Michael Segal and do not necessarily represent those of IDG Communications, Inc., its parent, subsidiary or affiliated companies.

More from this author