Advancements in Edge Computing: Applications and Benefits

Slide Note

Edge computing is revolutionizing the way we interact with technology by reducing latency and enhancing various applications such as augmented reality, predictive maintenance, and smart cities. This technology addresses the requirements of low latency, huge data bandwidths, and security, offering advantages like improved response times, bandwidth reduction, and energy efficiency. With its potential for proximity services and data sovereignty, edge computing is reshaping the future of communication and computation convergence.

ham_agc Follow

Uploaded on Oct 05, 2024 | 0 Views

Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

E N D

Presentation Transcript

Convergence of Compute and Communication Shailesh Kumar Interra Systems Inc.

Broadband Networks

Applications we want AR wearables (smart glasses) IoT for automation, predictive maintenance Asset tracking Driverless cars Vehicle to vehicle & vehicle to infrastructure Smart cities Surveillance Smart parking Emergency response High bandwidth media sharing on local social media Live event streaming Immersive Multi-Player Gaming

Requirements we face Low latency Huge data bandwidths Security Isolation Geo-fencing

Mobile Edge Computing

Cloud Vs Edge Courtesy: NTT

Edge Computing Applications Courtesy: NTT

Edge Accelerated Web Platform Courtesy: NTT

Mobile Computing Architectures Mobile Edge Computing Fog Computing Cloudlets

Fog Computing Courtesy: Wang et al, 2017

Cloudlets Courtesy: Wang et al, 2017

General Edge Networks Courtesy: Wang et al, 2017

Advantages of Edge Computing Reduced Latency Augmented reality Cognitive assistance 51% better response time Bandwidth Reduction Up to 67% for BW hungry applications Security Isolation Disaster resilience Energy Efficiency Up to 42 % Reduction Active time to Idle time Type of apps Proximity Services Device 2 Device Context Information Location based Reliability Data sovereignty Geofencing

Mobile Computing Architectures MCC 2 tiers High Cloud providers MEC 3 tiers Low Mobile operators Medium RAN FOG 3 or more tiers Low Fog node owners Small Between Device and Data Center CLOUDLET 3 tiers Low Home or Local Business Small Device or between Device and Data Center Maybe Hierarchy Latency Ownership Population Location Large Data centers Context awareness Cooperation No Yes Yes No No Yes No Courtesy: Wang et al, 2017

Applications

Network Function Virtualization Network Functions Routers Switches Firewalls Implemented in Software Run in VMs or Dockers On off the shelf hardware At Edge of network

Gaming Challenges Bandwidth Intensive Latency Sensitive Opportunity Co located gamers Game servers at edge Advantages Millisecond latency Thin client games Can run on low end phones with browsers Most computing on edge server Edge HW and SW maintained separately Side Effects Play More with Neighbors

Smart City Surveillance High Res surveillance cameras distributed over various fixed and mobile (wearable or drone) end points in the city Local edge servers receive images Face localization, key point extraction, feature vector construction at edge Feature vectors sent to central servers in cloud Match in nation wide database and generate alerts

Smart City Surveillance Courtesy: AWS

Augmented Reality Courtesy: AWS

Virtual Desktops Your virtual desktop environment in your office Its virtual machine migrates to the nearest edge site as you travel Low latency between you and your VDE The feel of office wherever you are! May be that s not what you really want :P

Video Conferencing Globally distributed edge fabric of voice and video servers Individual video/voice streams don t have to go to central cloud servers Low latency for people near same edge server. Resilient to partial network outages.

Build your own CDN No need to use say Akamai or Limelight Edge network providing caching services like compute and storage services on cloud Watch who is accessing your content Spin up a micro cache on demand in high usage area.

Edge Computing at Chick-Fil-A Forecast how many waffle fries to cook every minute of the day? Main forecast in cloud with data from many restaurants. Lacks accuracy: Misses local traffic spikes, events, sports, weather etc. Augment with real time data in restaurant From point of sale systems Micro-adjust the initial forecast for local factors