Daminda Dinesh

What and why of edge computing

Dinesh Imaduwa Gamage's photo
Dinesh Imaduwa Gamage
·

4 min read

Table of contents

What is edge computing?

An Edge network is a distributed computing framework with edge servers and devices. It brings enterprise apps closer to the data source. Edge computing Improves response time to get faster insight from data collected by IoT devices. This is because edge computing networks do not need a centralized cloud to function. However, as necessary edge networks can connect to the cloud and send valuable data to the cloud for storage.

Why do we need it?

  • Optimize Bandwidth At present, many IoT devices use a centralized, often private, cloud to send data, perform analytics and get insights to trigger actions. However, the increase of connected IoT devices will continue to grow. Consequently, the volume of data being streamed on networks will continue to grow. The existing network and infrastructure are not capable of coping with this demand. This is where edge computing is critical. As It can process data and filter data sent from the IoT devices closer to the devices and send the data that is not time-sensitive to the centralized data management platform, can optimize the bandwidth.

  • To address issues related to latency IoT devices collect data, analyze those data and take action based on the interpretation of the data(the insights generated). Some of these actions are time-sensitive and critical and should perform in real-time. Ex: — A self-driving car has to stop when a pedestrian suddenly crosses the street. The car must stop immediately. — A cooling system needs to turn on as soon as its sensors overheat.

  • The need for edge computing will continue to grow with 5G in use — 5G network promises greater speed and reduction in latency. Mobile edge computing and 5G network together can achieve ultra-low latency. This can potentially help in applications such as remote telesurgery or autonomous drones. — Sending data back and forth from the cloud can be expensive and deteriorate customer experience even with 5G, especially with data-heavy applications. As edge servers can filter out data and send what is necessary for the cloud for storage, this can improve customer experience and decrease the cost of connectivity in the networks.

  • Better security The potential for data compromise is minimized with edge computing because edge servers are closer to the data source and the distributed nature of edge networks. But, still, it is important to apply specific security measures to protect data against hackers.

What is an edge server?

  • Placed at a logical “edge” of a network
  • Serves as the connection between separate networks
  • Primary purpose: store content as close as possible to a requesting client machine to reduce latency and improve page load times.
  • Edge server belongs to a family of edge devices ( ex: routers, routing switches.)
  • These devices are placed inside internet exchange points to allow different networks to connect and share transit.
  • Although there is no clear definition of edge server placement, four main types can be identified
  • Device edge (the server close to the end-user device).
  • On-prem edge (physically located in the local network or facility).
  • Network edge (network-specific nodes such as base stations or telco data centers).

  • Regional edge (the server is in a local data center).

  • Cache static assets such as images, HTML and javascript files, and other static content.

Two distinct types of edge servers

1. Content delivery network (CDN) edge servers

  • Content caching
  • Provide users a cached version of static content( images, JS, HTML, downloadable content) from origin servers
  • Help reduce the workload on the origin server
  • Reduce latency for users
  • Deployed at points of presence (PoPs) and edge locations across a content delivery network.

2. Edge compute servers

  • Provide compute resources at the network edge
  • Strategically deployed to reduce latency
  • Provides functionality(data processing) for IoT applications and 5G networks
  • What is an Origin Server?
  • The main purpose of the origin server is to process and respond to incoming internet requests from clients. There can be latency in responding to client requests. This latency is attributed to
    • traffic exceeding the capabilities of the server
    • the physical distance between the client and the server
    • the additional round-trip time (RTT) for ensuring secure connection through SSL/TLS

Summary

An edge server sits between two distinct networks and is close to the data source Reduced latency and bandwidth optimization are the primary benefits of edge servers. There are mainly two types of edge servers CDN edge servers: decrease latency in static web content delivery Edge compute servers: decrease latency in data processing in cases such as IoT and 5G network

CDNcloudflareCloud Computingdistributed systemedgecomputing