Edge Computing
54 likes52,728 views
The document discusses the importance of edge computing, highlighting its applications in autonomous tasks, latency-sensitive operations, and significant bandwidth requirements. It emphasizes the market growth for edge computing and its benefits such as cost-effectiveness, security, and faster response times. However, it also points out the challenges related to managing edge deployments, including monitoring, maintenance, and operational complexities.
Edge Computing
- 1. Edge Computing IEEE ComSoc workshop on Evolution of Data-centers in the context of 5G Chetan Kumar S Co-founder and CEO, Aikaan Labs Pvt Ltd
- 2. WHY ?! Care about Edge Computing ?
- 3. Applications that require Autonomy ● Completing tasks with little or no human interaction – Self Driving Cars – Industry 4.0 ● It’s pretty obvious that ingesting data to a distant cloud a realistic option.
- 4. Applications that can’t tolerate Latency ● Latency can be reason for failure – Health care – Financial transactions ● It’s pretty obvious that ingesting data to a distant cloud isn’t a realistic option.
- 5. Applications that require significant Bandwidth ● Generate significant amount of data – Smart Surveillance system ● It’s pretty obvious that ingesting data to a distant cloud isn’t a realistic option.
- 6. What does market say ?
- 7. 81% Increase in revenue of edge computing by 2018 end North America (with 31 percent of revenue) and Europe (with 35 percent) will see the biggest jumps.
- 8. 72% The average anticipated annual revenue growth for IoT edge vendors by 2018 end
- 9. WHAT ?! Is Edge Computing ?
- 10. The Edge Computing ● Edge computing is optimization of cloud, to move the compute close to the source of data, to the edge
- 11. 11 Sensors and controllers Edge ● Real time data processing ● Real time control (M2M) ● Local data filtering and caching ● At source data visualization Cloud ● Central processing of summary data ● Big data analysis, complex learning model ● Central control IoT Gateway IoT Gateway Server Network Switch Storage Internet Gateway Cloud Apps The Edge Computing Internet Intra net Proximity
- 12. The Edge Computing Characteristics ● Geographically distributed ● Scalable ● Autonomous and distributed ● Edge computing is contextual and low latency ● Real-time interactions ● Heterogeneous
- 13. Benefits of Edge Computing 13 C ost Ef ective S ecu re Faster R esp on se R eliab le op eration s ● Locally stored ● No theft during transport ● Compliance maintained Secure ● Can work without connectivity. Reliable Operations ● Operating at the source of data ● Faster response time for triggers Faster Response ● No need to transport everything to cloud ● No recurring cost Cost Effective Edge Network
- 14. Challenges of Edge Deployments Not simple !
- 15. 1515 Post deployment, Monitoring and Managing Edge Infrastructure becomes a nightmare as deployments are: Open to environment Price sensitive/no redundancy Remotely locatedComplex Maintenance Criticality
- 16. Day-to-Day Edge Operational Challenges 16 Temperature sensor data received on time ? Is IOT network Overloaded ? No data after a rainy day; network down, sensor issue ? Traffic Lights and Camera not in Sync ..
- 17. Manageability Challenges 17 Where to start debugging from Truck rolls to the site Cloud tools lacks holistic view of complete Edge Infrastructure
- 18. Management and Control Model 18 IoT Devices IoT Edge Infrastructure Cloud IoT Gateway IoT Gateway Server Network Switch Storage Internet Gateway Edge Controller at cloud Edge Controller on premise
- 19. THANKS! Any questions? You can find me at chetansk@aikaan.io / twitter @chetansk www.aikaan.io Disclaimer: Acknowledge all owners of the images and video used in this presentation, used for knowledge sharing purpose, no commercial usage intended
- 20. Backup
- 21. The Fog, Edge and Mist Computing