Livin on the Edge, Part 4: Bringing Analytics & Business Logic to Edge Compute

This is part four of a five part series on edge computing with a focus on software and security.

THE COWEN INSIGHT

Edge computing targets use cases that are not well-addressed by the cloud, most often because of networking and latency constraints. By combining real-time data from edge devices with analytical & transactional processing, new types of intelligent applications will be built. Most edge software spending is for IIoT apps today, but we expect adoption from Enterprise IT to accelerate in the years ahead.

The Enterprise Computing Pendulum Is Starting to Swing

We have witnessed multiple enterprise computing cycles over past decades, from centralized mainframe computing in the‘60s/’70s to de-centralized client/server computing in the ‘80s/’90s and back to centralized computing via the cloud in the ‘00s/’10s. We think we are on the cusp of computing innovation swinging back to distributed architectures via edge computing as we lookout over the next couple of decades.

Indeed, the innovation & cost curve coming from processors, sensors, wireless bandwidth and open-source software sets the stage for artificial intelligence, machine learning, analytics & enterprise applications to be powered by edge networks & devices. These help form a new set of modern applications that are intelligent & automated. We think hybrid strategies will come to include data center, cloud and now edge computing.

Edge Computing Augments Software Development

As telcos and cloud providers roll out new computational, storage & networking infrastructure in edge locations, there will be a new generation of computing capabilities for software developers to tap into. This comes at a time when the cost of collecting wireless data is falling and the innovation of artificial intelligence and machine learning algorithms is accelerating.

Developers can take advantage of ultra-low data processing latency and localized application logic that can power intelligent applications, machine-to-machine automation, real-time visibility & analytics, immersive applications and more.

Industrial IoT captures most of today’s edge revenue; Enterprise IT is nascent

Most of today’s edge computing software spend is with vertically-designed Industrial IoT applications. This includes hardware (sensors, cameras, servers), asset connectivity (wireless/wireline communication protocols), data storage & processing (edge-to-cloud), advanced analytics (artificial intelligence, machine learning), advanced monitoring & diagnostics, and edge application delivery (including operational intelligence solutions that power next-gen Smart applications).

The opportunity for edge to make its way into Enterprise IT is still very early. We’ve found very few horizontal software vendors selling into Enterprise IT or carrying much of a dedicated roadmap around edge computing. Many have IoT solutions, being addressed through applications run in the Cloud. There is certainly a compelling revenue opportunity, but very few have major initiatives to support deploying applications directly at the edge to date. However, as more technology standards get established, ROIs of new investments get clearer. As IT gets comfortable with security, we expect to see a new generation of software being built.

Killer apps are smart, autonomous & immersive

Edge computing reduces reliance on network bandwidth and lowers throughput latency due to near instant compute & analytics. A reduced reliance on network bandwidth and lower throughput latency is especially powerful for applications that need <100 milliseconds of data processing throughput. This accelerates artificial intelligence and machine learning based decision-making. It also eliminates processing lag time for things like Smart applications (i.e. predictive maintenance, smart retailing, smart cities), autonomous applications (i.e. cars, drones, industrial equipment) and immersive applications (i.e. augmented reality, virtual reality, medical-related computer-vision, wearables). Essentially, any application workload that can be powered by real-time connected data, whether a Smart system or an artificial intelligence or machine learning algorithm, is well-suited for being deployed in an edge architecture.


Read the five part edge computing series

Part 1: Evolving Tomorrow’s Internet focuses on communications services and cloud/internet.

Part 2: The Brains and Nervous System of Edge Computing focuses on computing and memory.

Part 3: Storage & Networking focuses on storage and networking.

Part 4: Bringing Analytics & Business Logic to Edge Compute focuses on software and security.

Part 5: Enabling & Empowering Locally focuses on end-devices & services enabled by edge computing.

Get the Full Report

If you’re already a member of our Research portal, log in.

Log In

If not, reach out to us directly for more information.

More Like This

Concept around the premise that robotics is a key tool for combating climate change, fostering sustainable industry, and supporting ESG. The image is of two assembly robots making a digital rendering of a tree, effectively growing and nurturing the tree. Neon blue lighting against a black background with a city-scape visible in the background.
Ahead of the Curve®

Deus Ex Machina Part IV, Robotics as Tools in Climate Fight

Read More
Podcast
Concept for reservoir analysis used to evaluate CO2 storage. Related to sustainability, clean energy, and big data. The image is of a light bulb growing in soil with data circuits circulating around the lightbulb as if reflected by yellow light of the bulb.

Carbon Storage Evaluation with Core Lab

Read More
Video

50 Years of Innovative Ideas: Cowen’s TMT Conference

Read More