Edge Analytics in IoT Image

Edge Analytics in IoT


Whether the intelligence of IoT systems and subsystems should reside at the cloud or at the edge is one of the most critical questions in the industry. While the historical focus favored cloud platforms for their ability to create highly sophisticated ML-based algorithms, a greater emphasis on edge analytics is on the horizon with advances in device computing capabilities, as well as the enabling connectivity infrastructure, creating a growing number of instances in which it makes more sense to perform analytics closer to the ‘thing’ or activity that is generating or collecting data. These small footprint edge analytics solutions allow companies the freedom and flexibility to differentiate their products and services. They also allow machines and connected equipment to continuously self-regulate, equip engineers and industrial process analysts with an outlet to innovate around inefficiency, and offer a vantage point for procurement professionals to improve the products and services they provide as a result of greater visibility throughout the supply chain.

This research analysis examines what ABI Research considers to be the most significant trends and developments related to edge analytics in IoT. The first section includes the conceptual framework supporting the research, as well as a wrap-up of recent activity. The second section consists of updated and refined forecasts on the market’s growth, broken down by application segment, connectivity technology (2G, 3G, 4G, Fixed Line, Satellite), and revenue potential. The final section provides a high-level assessment of a number of relevant vendors to exemplify the different parts of the value chain. As a whole, this analysis should be viewed as an extension of ABI Research’s earlier work around IoT analytics, which evolved considerably since the initiation of coverage in 2013. Recent examples include “Big Data Analytics in IoT and M2M” (AN-2255), “Machine Learning in IoT” (AN-2016), and “Big Data and Machine Learning in Telecom” (AN-2302).

Table of Contents

    • 1.1. The Analytics Value Chain in Review
    • 1.2. Edge Computing vs. Edge Analytics
    • 1.3. IoT Intelligence: from Endpoint to Cloud
    • 1.4. Drivers and Inhibitors for Edge Analytics
    • 1.5. Key Trends and Observations
    • 2.1. Methodology
    • 2.2. Application Segmentation
    • 2.3. Volume of Generated IoT Data
    • 2.4. Volume of Captured IoT Data
    • 2.5. Volume of Transmitted IoT Data
    • 3.1. AGT International
    • 3.2. Bit Stew Systems (GE)
    • 3.3. Camgian Microsystems
    • 3.4. Cisco
    • 3.5. CyberLightning
    • 3.6. Dell
    • 3.7. Intel
    • 3.8. Foghorn Systems
    • 3.9. Kepware Technologies (PTC)
    • 3.10. Mtell
    • 3.11. Osisoft
    • 3.12. Panduit
    • 3.13. Parstream (Cisco)
    • 3.14. Predixion Software (Greenwave Systems)
    • 3.15. Resin.io
    • 3.16. Sight Machine