Edge Computing
Edge computing means running software and processing data physically close to where it is created, on a factory sensor, a store's local server, or a cell tower, instead of sending everything to a distant cloud data center first. It exists because some tasks need an answer faster than that round trip allows, or must keep working when the network connection drops.
itCloud computing | OpenSkills.info
Course pathWalk it in order
Look it upDip in anytime
Go furtherLeaves this page
Don't Panic
Don't Panic - Edge Computing
Edge computing places processing and storage close to where data is produced or consumed, instead of sending every byte to a distant cloud region first. It complements cloud computing. Most designs still use a central cloud for aggregation, long-term storage, training, and fleet management. The edge tier handles work that cannot wait, cannot afford the bandwidth, or must continue when the link is poor.
The Linux Foundation Open Glossary and NIST's fog computing model describe the same shift: cloud-only paths struggle with device scale, variety, and delay. In practice you meet three overlapping flavors. Device edge runs on or beside the machine producing data. Infrastructure edge places smaller regional or on- premises capacity between devices and hyperscale regions. Network edge embeds compute in CDN or telecom locations closest to end users.
Choose placement from a concrete constraint. A control loop, a saturated uplink, or an unreliable site link is a reason. A marketing label is not. Expect operations to dominate: provisioning, secrets, updates, observability, and offline behavior are harder on constrained fleets than in a single region.
Keep the mental model simple. Move the decision that has the constraint. Leave the rest where management and durable storage are cheaper. Measure whether the local path actually removed latency, bandwidth, or availability pain.
Read the Intro for the three edge flavors and constraint drivers. Use the Cheatsheet when you need the placement checklist. Landscape places edge platforms among related distributed-systems options; Updates tracks NIST SP 500-325, the conceptual model this course uses beside vendor examples.
Where this skill leads
Relevant careers
See how this topic contributes to broader role-level skill maps.
Sources
- https://github.com/State-of-the-Edge/glossary/blob/master/edge-glossary.md
Supports
- Edge computing is the delivery of computing capabilities to the logical extremes of a network in order to improve the performance, operating cost and reliability of applications and services
- Device edge is edge computing capability on the device or user side of the last mile network, often depending on gateways to collect and process data from devices
- Fog computing is an early edge computing concept that stipulates compute and data storage resources, as well as applications and their data, be positioned in the most optimal place between the user and cloud
- A cloudlet is a mobility-enhanced public or private cloud at the infrastructure edge, used interchangeably with edge cloud, edge data center, or edge node
- Data gravity is the concept that data is not free to move over a network and that the cost and difficulty of doing so increases as both the volume of data and the distance between network endpoints grows
- Data sovereignty is the concept that data is subject to the laws and regulations of the country, state, industry it is in, or the applicable legal framework
- https://csrc.nist.gov/pubs/sp/500/325/final
Supports
- Cloud-based approaches to IoT struggle with the massive scale, variety of devices, and substantial delays involved in centralizing processing
- Fog computing decentralizes applications, management functions, and data analytics into the network itself, using a distributed and federated compute model
- The publication characterizes how fog and mist computing architectures complement and differ from traditional cloud-based systems for IoT deployments, and addresses related concepts including edge computing and cloudlets
- https://aws.amazon.com/what-is/edge-computing/
Supports
- Edge computing is the process of bringing information storage and computing abilities closer to the devices that produce that information and the users who consume it
- The unprecedented complexity and scale of data have outpaced network capabilities
- Key benefits include reduced latency, improved security, increased productivity, remote data collection, reduced costs, and reliable performance in areas with scarce connectivity
- Industries using edge computing include manufacturing (IoT automation), autonomous vehicles (real-time decision-making), energy (pipeline monitoring), and healthcare (patient data privacy)
- Cloud service providers also provide edge computing services, and 5G networks enable ultra-low-latency applications at the network edge
- https://learn.microsoft.com/en-us/azure/iot-edge/about-iot-edge
Supports
- Azure IoT Edge is a device-focused runtime that enables you to deploy, run, and monitor containerized Linux applications, bringing analytics closer to devices for faster insights and offline decision-making
- You can run anomaly detection at the edge to respond as quickly as possible to emergencies happening on a production line
- Azure IoT Edge is made up of three components — IoT Edge modules, the IoT Edge runtime, and a cloud-based interface for remote monitoring and management
- IoT Edge runs natively on Linux and on Windows devices through IoT Edge for Linux on Windows
- https://kubeedge.io/
Supports
- KubeEdge is a Kubernetes Native Edge Computing Framework that extends containerized application orchestration to edge devices
- KubeEdge reduces transmission costs and shrinks latency between the data and the decision
- KubeEdge supports bidirectional communication and synchronization of metadata and data between cloud and edge, including edge nodes in private subnets
- KubeEdge supports autonomous operation of the edge even during disconnection from the cloud, with metadata persisted per node
- KubeEdge has a memory footprint down to approximately 70 MB
- https://developers.cloudflare.com/workers/
Supports
- Cloudflare Workers is a serverless platform for building, deploying, and scaling apps across Cloudflare's global network with no infrastructure to manage
- Workers executes code across Cloudflare's distributed global network so applications run closer to end users geographically, delivering fast performance with high reliability anywhere in the world
- https://aws.amazon.com/edge/
Supports
- AWS Wavelength offers compute and storage services with the same developer experience as in the cloud, embedded within communication service providers' data centers at the edge of the 5G network
- AWS Snow Family devices are suitcase-sized, rugged devices designed to run local computing applications and migrate data into the cloud in non-data-center environments with limited or no connectivity
- AWS extends infrastructure, services, APIs, and tools offered in the cloud as a fully managed service to virtually any on-premises data center or co-location space
- https://www.cncf.io/blog/2021/12/08/hosted-edge-vs-cloud-the-battle-for-latency-and-security/
Supports
- A hosted edge with distributed nodes geographically close to users can substantially reduce latency compared to centralized cloud data centers concentrated in urban areas
- A hosted edge is harder to secure than a core network and more variable than the big public clouds
- Storing data in multiple third-party networks is risky because of the threat of data leaks, localized intrusions, and different security and compliance standards in different regions
- A web application firewall running at the edge can secure critical applications and data
- Choose hosted edge to prioritize latency for users in remote locations and real-time responsiveness; choose cloud to prioritize consistent security posture and simplified infrastructure management
- https://lfedge.org/projects/
Supports
- Akraino provides a set of open infrastructures and application blueprints for the edge, spanning use cases including 5G, AI, Edge IaaS/PaaS, and IoT
- EdgeX Foundry is a flexible and scalable open source edge platform that facilitates interoperability between devices and applications at the IoT edge
- Project EVE creates an open Linux-based operating system for distributed edge computing infrastructure