Edge & IoT — Trivia & Interesting Facts¶
Surprising, historical, and little-known facts about edge computing and IoT infrastructure.
The term "Internet of Things" was coined in 1999 by a Procter & Gamble employee¶
Kevin Ashton, a brand manager at Procter & Gamble, coined the term "Internet of Things" in a 1999 presentation about using RFID tags to track products through supply chains. He later said he chose "Internet" in the title because it was the hottest buzzword of 1999 and would get attention. It worked.
There are over 15 billion IoT devices worldwide, outnumbering humans 2 to 1¶
By 2024, the number of connected IoT devices exceeded 15 billion, roughly double the world's population. Projections suggest this will reach 30 billion by 2030. Most of these devices run minimal operating systems and have severe resource constraints — some have as little as 256KB of RAM.
Edge computing predates cloud computing¶
Content Delivery Networks (CDNs) like Akamai, founded in 1998, were doing "edge computing" before the term existed — caching content at points of presence near users to reduce latency. The modern edge computing movement essentially extends this concept from content caching to general-purpose computation.
Kubernetes was not designed for edge, but K3s made it work¶
Rancher Labs created K3s in 2019 — a Kubernetes distribution packaged as a single binary under 100MB. Standard Kubernetes requires several gigabytes of RAM and multiple components. K3s made Kubernetes viable on Raspberry Pis, industrial gateways, and retail point-of-sale systems, opening edge deployments to the Kubernetes ecosystem.
AWS Snowball Edge is literally a computer shipped in a box¶
AWS Snowball Edge is a ruggedized hardware device that AWS physically ships to customers for edge computing and data transfer. It contains compute, storage, and even GPU capabilities. For customers in remote locations or with massive datasets, it's faster to FedEx a Snowball than to transfer data over the network.
The average autonomous vehicle generates 5-20 TB of data per day¶
Self-driving cars generate between 5 and 20 terabytes of data per day from cameras, LIDAR, radar, and other sensors. This data cannot be sent to the cloud for processing in real-time — the latency would make the car dangerous. Edge computing in vehicles must process this data locally in milliseconds.
MQTT was invented for monitoring oil pipelines in 1999¶
The MQTT protocol, now the dominant IoT messaging protocol, was created by Andy Stanford-Clark (IBM) and Arlen Nipper in 1999 for monitoring oil pipeline sensors via satellite. It was designed for unreliable networks with limited bandwidth — constraints that perfectly matched IoT devices two decades later.
Edge locations now process more data than centralized clouds¶
By 2025, Gartner estimated that 75% of enterprise-generated data would be created and processed outside traditional centralized datacenters. The volume of data generated at the edge — from sensors, cameras, vehicles, and industrial equipment — has made it impractical to backhaul everything to the cloud.
A compromised IoT device was used to attack infrastructure in 2016¶
The Mirai botnet, discovered in 2016, hijacked approximately 600,000 IoT devices (IP cameras, DVRs, routers) using default passwords and used them to launch a DDoS attack against Dyn DNS, taking down Twitter, Netflix, Reddit, and dozens of other major websites. It demonstrated that IoT security is infrastructure security.
WebAssembly is emerging as the edge runtime of choice¶
WebAssembly (Wasm), originally created for web browsers, is being adopted as an edge computing runtime because it starts in microseconds (vs. milliseconds for containers), uses minimal memory, and provides sandboxed execution. Companies like Fastly and Cloudflare run Wasm-based edge compute platforms that handle millions of requests per second.