Introduction to

Edge Computing
Ken-Hao Chang
Taiwan Evolutionary Intelligence Laboratory
2017/11/12 Group Meeting Presentation
Outline
• Trend of Computing
• Cloud Computing
• Edge Computing
• Application
• EdgeX project
• AWS lambda@edge
Trend of Computing

5th Utility
Cloud Computing
• Cloud computing is a distributed technology which delivers hosted
services over the internet to provide easy access, scalable (and
elastic) IT services
Five characteristics
1. On-demand Self-service
A consumer can unilaterally provision computing capabilities, such as server time
and network storage, as needed automatically without requiring human
interaction with each service’s provider.
Five characteristics
2. Broad Network Access
Capabilities are available over the network and accessed through
standard mechanisms that promote use by heterogeneous client
platforms
Five characteristics
3. Resource Pooling
The provider’s computing resources are pooled to serve multiple
consumers using a multi-tenant model, with different physical and
virtual resources dynamically assigned and reassigned according to
consumer demand.
Five characteristics
4. Rapid Elasticity
Capabilities can be rapidly and elastically provisioned, in some cases
automatically, to quickly scale out, and rapidly released to quickly scale
in.
Five characteristics
5. Measured Service
Cloud systems automatically control and optimize resource use by
leveraging a metering capability at some level of abstraction
appropriate to the type of service .
Benefits From Cloud
• For the end user and individuals
• Improved performance
• Instant update & free software
• Device independence & Mobility
• For the market and enterprises
• Reduce capital expenditure
• Improve Business Agility & Specialization
• Fast Deployment & Continuous Service Delivery
Cloud Computing structure
Privacy protection
an obstacle for cloud computing in IoT

Data quantity is too large

Transportation is becoming the bottleneck
Emergence of Edge Computing
• Mobile computing
• Internet of Things( IoT )
• Smart Home : smart light, smart TV, and robot vacuum
• Just connecting to cloud is not enough
• Augmented Reality (AR)
What is Edge Computing
• Edge
• Any computing and network resources along the path between data sources
and cloud data centers

Edge

Data Source
Three-level Architecture
• First level:
• unmodified cloud infrastructure
• Second level:
• dispersed elements called cloudlets
with state cached from the first level
• Third level:
• Mobile-device or IoT device
Technology
• Content Delivery Network (CDN)
• Fog nodes
• Cloudlets
• Micro datacenters
CDN
• Geographically distributed network of proxy servers and their data
centers
• Benefits
• Reducing bandwidth costs
• Improving page load times
• Increasing global availability of content
• Location-relevant advertising
• Techniques
• web caching
• server-load balancing
• request routing
Cloudlets ( Edge )
• A mobility-enhanced small-scale cloud datacenter that is located at
the edge of the Internet

Cloudlet Cloud Cloudlet

Cloudlets
• Based on standard cloud technology (virtual machines ) : Openstack
• Soft state only ( Storage )
• contain cached states from the cloud
• buffer data originating from a mobile device and going to the cloud
• entirely self-managing
• Close at hand
• “Logical proximity" of the mobile devices
• Resources and connectivity (Networking and Processing)
• sufficient CPU, RAM, etc. to offload resource-intensive computations from several mobile
devices
• good connectivity ( bandwidth) to the cloud
• not limited by electric power supply
Processing

Raw Data

Result
Cloudlet
Processing

Raw Data Extracted Data

Result Result
Cloudlet
Cloud
Benefits of Cloudlets
• Highly responsive cloud services.
• low end-to-end
• Low latency, high bandwidth
• Scalability via edge analytics.
• Raw data is analyzed on cloudlets. Only the extracted information transmitted to the
cloud.
• Privacy-policy enforcement.
• A cloudlet can enforce the privacy policies of its owner prior to release of the data to the
cloud
• Masking cloud outages.
• If a cloud service becomes unavailable , a fallback service on a nearby cloudlet can
temporarily mask the failure
Compare to Cloud Computing
Parameters Cloud Computing Fog Computing
Server nodes location Within the Internet At the edge of the local network
Client and server distance Multiple hops Single/multiple hop
Latency High Low
Delay Jitter High Low
Security Non-locally controllable Locally controllable
Location awareness No Yes
Vulnerability Higher probability Lower probability
Geographical distribution Centralized Dense and Distributed
Number of server nodes Few Very large
Real time interactions Not fully supported Supported
Usual last mile connectivity Leased line /wireless Mainly wireless
Mobility Limited support Supported
Challenges
• Collective control and sharing of cloudlets in distributed computing
• Managing dispersed cloudlet infrastructure
• Higher management cost than centralized infrastructure
Edge X
• Vendor-neutral open source software platform at the edge of the network
• Interacts with the physical, every day working world of devices, sensors,
actuators, and other IoT objects.
• The EdgeX platform enables and encourages the rapidly growing
community of IoT solution providers to work together in an ecosystem of
interoperable components to reduce uncertainty, accelerate time to
market. 
• Easier to
• monitor / send instructions / collect data from physical world items
• move the data across the fog up to the cloud where it may be stored, aggregated,
analyzed, and turned into information, actuated, and acted upon
AWS Lambda@Edge
Run Lambda functions at AWS Regions and Amazon CloudFront edge locations
Reference
• Wearable Cognitive Assistance using Cloudlets (
http://elijah.cs.cmu.edu/)
• The Emergence of Edge Computing, Computer 50.1 (2017): 30-39.
• Edge Computing: Vision and Challenges, IEEE INTERNET OF THINGS
JOURNAL, VOL. 3, NO. 5, OCTOBER 2016
• K.P.Saharan A.Kumar “ Fog in Comparison to Cloud: A Survey”, Int’l.
Journal of Computer Applications (0975 – 8887) Volume 122 – No.3,
July 2015