z6110x0035: introduction to cloud computing - cloud edge...
TRANSCRIPT
![Page 1: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/1.jpg)
Z6110X0035: Introduction to Cloud Computing
- Cloud–Edge Computing
Lecturer: Prof. Zichen Xu
Acknowledge Dr. Weisong Shi, ApproxIoT, and Tsinghua University
![Page 2: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/2.jpg)
Recap from Previous Class
Finals!
![Page 3: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/3.jpg)
Cloud-edge Computing Paradigm
物联网 云计算
![Page 4: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/4.jpg)
Why do we need edge computing
Push from cloud services
Pull from IoT
Change From Data Consumer to Producer
![Page 5: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/5.jpg)
What is edge computing
We define “edge” as any
computing and network
resources along the path
between data sources and cloud
data centers.
Edge computing is
interchangeable with fog
computing.
![Page 6: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/6.jpg)
Case study
1. Cloud offloading (online shopping services)
2. Video analytics (finding a lost child in the city)
3. Smart home
4. Smart city
5. Collaborative edge
![Page 7: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/7.jpg)
Challenges and opportunities
ProgrammabilityComputing stream that is defined as a serial of functions/computing applied on the data
along the data propagation path.
The function/computing distribution metric could be latency-driven, energy cost, TCO,
and hardware/software specified limitations.
![Page 8: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/8.jpg)
Challenges and opportunities
Naming
![Page 9: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/9.jpg)
Challenges and opportunities
Data abstraction
![Page 10: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/10.jpg)
Challenges and opportunities
Service Managementdifferentiation, extensibility, isolation, and reliability
Privacy and Security
Optimization Metrics(latency、bandwidth、energy、cost)
![Page 11: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/11.jpg)
Quantifying the Impact of Edge Computing on
Mobile Applications
How much can edge computing actually benefit end users
Latency and energy consumption
Compute-intensive and latency-sensitive applications such as
mobile augmented reality
![Page 12: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/12.jpg)
Experimental applications
Comet: a existing tool that can transparently migrate threads
from a mobile device to a remote server and back
![Page 13: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/13.jpg)
Experimental Setup
Cloudlet: a mobility-enhanced small-scale cloud datacenter
that is located at the edge of the Internet.
![Page 14: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/14.jpg)
Experimental Setup
No offload
Cloud-WiFi
Cloudlet-WiFi
Cloudlet-LTE
Cloud-LTE
![Page 15: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/15.jpg)
WiFi offloading performance
![Page 16: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/16.jpg)
WiFi offloading and energy
![Page 17: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/17.jpg)
LTE offloading performance
![Page 18: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/18.jpg)
Mobile Edge Computing: Progress and Challenges
![Page 19: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/19.jpg)
A Edge Analytics Demo
IBM
![Page 20: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/20.jpg)
An Edge Analytics Demo
This demo is to showcase the following
1.How sensors and digitized elements get locally connected with one or more IoT gateway instances in order to gather and transmit
any useful and usable data to the IoT gateway. In other words, multi-structed and massive data getting generated by various sensors
and sensors-attached assets in a particular environment (say, homes, hotels, hospitals, etc.) are received and temporarily stocked by
IoT gateways / middleware/brokers for purpose-specific data analytics.
2.By deploying an edge analytics and application development platform in the IoT gateway (Raspberry Pi was used for our demo), all
kinds of data getting collected are getting cleansed and crunched in real-time in order to emit out actionable and timely insights.
3.The IoT gateway also contributes in filtering out irrelevant data at the source itself so that a very limited amount of useful data gets
transmitted to the faraway clouds to facilitate historical and comprehensive big data analytics. The IoT gateway acts as an
intermediary between scores of on-premise edge systems and off-premise clouds.
4.IoT gateway modules (typically touted as fog devices) act as the master node/leader in monitoring, measuring and managing
various dynamic edge devices and their operational parameters
5.IoT gateway modules seamlessly and spontaneously integrate the physical world with the cyber world (cloud services, applications,
databases, platforms, etc.)
6.IoT gateway activates, augments, and adapts actuation devices (edge) based on the insights extricated through analytics in real
time
![Page 21: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/21.jpg)
Cloud
Edge Compute(Raspberry Pi)
The Macro-level Architecture of our Demo System
21
Edge Analytics(Edgent)
Container Management (Kubernetes)
Me
ssag
e B
roke
r (K
afka
)
Machine Learning / AI
Sen
sors
/Dev
ice
Co
ntr
olle
rs
Dev
ice
s
![Page 22: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/22.jpg)
The Demo Components
•Raspberry Pi Configuration Steps:
•https://www.raspberrypi.org/documentation/configuration/
•Model 3 b+, Configuration – 1 GB RAM, 64GB SD card
•Processor Type: Broadcom BCM2837B0, Cortex-A53 64-bit SoC @ 1.4 GHz
•Ports: 3 USBs, HDMI, 2 WLAN, 1 Ethernet, Bluetooth
•IR/Motion Sensor / Pulse Rate Monitor
•Pi4j - http://pi4j.com/download.html
•Apache Edgent 1.2.0
•https://developer.ibm.com/recipes/tutorials/setting-up-apache-edgent-on-my-raspberry-pi-3/
•Docker Container - through the clustering of heterogeneous edge / fog devices
•AWS Compute Instance
•Apache Flink 1.4.2
•https://data-flair.training/blogs/install-configure-apache-flink-ubuntu/
•Not used for this demo/workshop:
•Kafka
•Kubernetes
![Page 23: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/23.jpg)
Confidential | DD.MM.YY | version #
The Raspberry Pi PIN Layout
![Page 24: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/24.jpg)
Modern online services
Stream aggregator
Stream analytics system
UsefulInformation
Processing streaming data from different sources
![Page 25: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/25.jpg)
Modern online services
Low latency
Tension
Approximate computing
Efficient resourceutilization
![Page 26: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/26.jpg)
Approximate computing
Many applications:Approximate output is good enough!
The proportion of data is useful for this application
Live taxi heatmap
![Page 27: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/27.jpg)
Approximate computing
Idea: To achieve low latency, compute over a sub-set of data items instead of the entire data-set
Analyze
Approximate output ±error bound
Approximatecomputing(sampling)
![Page 28: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/28.jpg)
State-of-the-art system
StreamApprox [Middleware’17]
Approximate output ± error boundStreamApprox
Stream aggregator
S1
S2
Sn
…
Data stream
Cloud datacenter
Limitations: • It wastes bandwidth• It utilizes only cloud datacenter resources
![Page 29: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/29.jpg)
Edge computing
Cloud
Gateway
Edge nodeLocal processing
Source of data
Allows data to be processed at the edge node before it’s sent to the cloud
Opportunities: • Providing more computing resources • Saving bandwidth
![Page 30: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/30.jpg)
Edge infrastructure
Source: https://peering.google.com/#/infrastructure
Azure IoT edge
Watson IoT
AWS IoT
![Page 31: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/31.jpg)
Problem statement
To build a stream analytics systemBy utilizing the cloud and edge computing resourcesBy leveraging approximate computing
Design goalsEfficiency: Efficient utilization of computing resourcesAdaptability: Adaptive execution based on the available resourcesTransparency: No code change required and resource management
![Page 32: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/32.jpg)
ApproxIoT: Overview
S1
Si
Sn
…
Sm
……
Central node
Cloud
Query
Approximate output ± error bound
ApproxIoTApproxIoT employs sampling in the distributed environment of
edge + cloud
Edge nodes
Regional edge
Continental node
![Page 33: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/33.jpg)
Naïve algorithm
SRS Query
Simple random sampling (SRS)
Approximate output ± error bound
Sampled unfairly Overlooked Low accuracy
![Page 34: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/34.jpg)
Background: Stratified sampling
Stratified sampling
Advantage: The sub-streams are sampled fairly
Disadvantage: Requires the knowledge of each sub-stream size
![Page 35: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/35.jpg)
Background: Reservoir sampling
Reservoir sampling
Size of reservoir = 4
Reservoir sampling
Size of reservoir = 4
Advantage: • No pre-knowledge required of sub-stream size
Disadvantages: • The sub-streams are sampled unfairly• Difficult to run on multiple nodes
Reservoir sampling
Size of reservoir = 4
The 5th item With probability(4
5) replaced by the 5th item
Reservoir sampling
Size of reservoir = 4
Reservoir sampling
Size of reservoir = 4
The 6th item With probability(4
6) replaced by the 6th item
Reservoir sampling
Size of reservoir = 4
Reservoir sampling
Size of reservoir = 4
![Page 36: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/36.jpg)
ApproxIoT sampling algorithm
Easy to parallelize, requires no synchronization between
sub-streams
Weighted hierarchical sampling (WHS)
Combining stratified and reservoir sampling
Weight: C/N, if C>N 1, if C <=N
WHS
Reservoir size N=4 With initial weight 1
W=1W=1
W=1
W=6/4W=1
W=1
C=6
![Page 37: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/37.jpg)
WHS on edge nodes
Regional edge WHS
W=1W=1
W=1
W=6/2=3W=4/2=2
W=1
Continental node WHS
W=4W=1
W=3
W=4*5/2=10W=1*3/2=3/2
W=3
Reservoir size equals 2
Central node
CloudEdge nodes
Regional edge Continental node
Easy to parallelize, requires no synchronization between
computing nodes
Carried weight Current weight
![Page 38: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/38.jpg)
ApproxIoT in the cloud
Reservoir size equals 1
Query (sum)WHS
The weights are carried W=4/3*6/1 =8
W=1*4/1=4
W=1*2/1=2
± error bound
8* +4* +2*
W=4/3
W=1
W=1
Approximate output:
Central node
CloudEdge nodes
Regional edge Continental node
![Page 39: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/39.jpg)
Implementation
S1
S2
Sn
…
Kafka cluster
Streampub/sub
Edgenodes
Clouddatacenter
Data stream
Sampleddata stream
Sampleddata stream
See the paper for more details
Kafka Streams
![Page 40: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/40.jpg)
Experimental setup
Evaluation questionsAccuracy vs. sample sizeThroughput vs. sample size
Testbed: 25 nodes15 nodes for ApproxIoT deployment 10 nodes for Kafka cluster
Datasets: Synthetic: Poisson and Gaussian distributionReal: Brasvo pollution and New York Taxi Ride
See the paper for more results!
![Page 41: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/41.jpg)
Accuracy vs. sample size
0
20
40
60
80
10 20 40 60 80
Acc
ura
cy lo
ss(%
)
Sampling fraction(%)
SRS ApproxIoTLower
the better
ApproxIoT: ~2600X higher accuracy over SRS
The average is 0.035%
![Page 42: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/42.jpg)
Throughput vs. sample size
0
40
80
120
10 20 40 60 80 90 100
Thro
ugh
pu
t(k)
item
s/s
Sampling fraction(%)
Native SRS ApproxIoTHigher
the better
• ApproxIoT has low overhead compared to the native execution• ApproxIoT has similar throughput as SRS
![Page 43: Z6110X0035: Introduction to Cloud Computing - Cloud Edge Computinggood.ncu.edu.cn/~xuz/cc/L10.pdf · 2019-05-14 · Edge computing is interchangeable with fog computing. Case study](https://reader034.vdocuments.mx/reader034/viewer/2022042303/5ece386a00bbfb522729ad8a/html5/thumbnails/43.jpg)
Conclusion: The promise of edge computing
Infrastructure and design
Measurement works
Related applications The Design and Implementation of a Wireless Video Surveillance System