The Cloud

Overview
● What is the Cloud?
● Cloud options
What is the Cloud?
● Software runs on the internet instead of running on your computer
○ Google Docs
○ Web Mail

● Custom software can be created without worrying about resources

○ On Premise
- 70% Time to manage hardware & 30% Develop Business logic
○ Cloud Infrastructure (Flips this)
- 30% Time to manage hardware & 70% Develop Business logic
Embedded vs Cloud

http/json
REST API
Virtual Machine Emulation of a computer
Flavors of Cloud Offerings
● IAAS - Infrastructure As A Service

● PAAS - Platform As A Service

● SAAS - Software As A Service

SAAS - Software as a Service

Microsoft Office Installed on your PC (On Premise)

Vs

Google Docs on the Cloud ( SAAS)

IAAS - Infrastructure as a Service
● Virtual machine , storage and networking capabilities are hosted by
a service provider and offered to customers on-demand

● Customers can provision the hardware via Web or SSH

PAAS - Platform as a Service
● Lies between SAAS & IAAS Models

● Hardware + OS + Middleware + DB services are provided

Flavors of Cloud Offerings
● IAAS - Infrastructure As A Service
- Rent Kitchen Equipment like stove & oven
- You buy Ingredients & Cook

● PAAS - Platform As A Service

- Rent Kitchen Equipment & Ingredients
- You cook

● SAAS - Software As A Service

- You just eat out
Advantages
● Pay based on usage

● No capital costs

● Elastic scaling, On demand

● Focus on business logic and not infrastructure

Google vs Microsoft vs Amazon
● Google Cloud Platform
○ Slightly more capabilities on Big Data and Machine Learning
○ Cheaper (minute based!)
● Microsoft Azure
○ Focused on Microsoft products
● Amazon AWS
○ API’s are most evolved
○ Has the most services. No SAAS
○ Documentation and community is most mature
We picked Amazon AWS
● Free Tier Available in most countries

● Offers an iOT SDK

● You are free to choose your own platform

Amazon Snowmobile - ExaByte
Google Cloud Vision
● https://cloud.google.com/vision/
Summary
● No reason to run your own servers anymore

● Learn to leverage the power of the cloud services

● Hands on projects in this course based using AWS

AWS Services
Important Services

EC2 S3 RDS

Elastic Cloud Compute Simple Storage Service Relational Data Service

 Compute Resources
EC2 Virtual Machine

ELB Load Balancer

Lambda Modular Service
EC2 Container Service Docker
 Databases
RDS Relational Database
Management System
(RDBMS)

DynamoDB NoSQL database

ElastiCache Memcached

S3 FTP Server
 DevOps
CodeCommit Git

CodeDeploy Jenkins

CodePipeline Continuous Integration

OpWorks Orchestration (Chef)

 Big Data
Elastic MapReduce MapReduce

Data Pipeline Spark

Kinesis Kafka
Takeaway or Resources
● Aws has a multitude of services that anyone can take advantage of
● https://www.expeditedssl.com/aws-in-plain-english
Systems Architecture
Overview
● Basic Architecture overview
● IOT-based Architecture
Web Architecture
Web Architecture
IoT
Edge Computing
● Instead of all devices connecting to one
cloud, multiple edge clouds manage
devices
● Each edge is in charge of its own devices
● Offloads compute power to the edge
● Edge brings lower latency to high
priority tasks
● Cloud handles low priority, but compute
intensive tasks.
Edge Computing
Takeaway or Resources
● https://aws.amazon.com/architecture/
● https://s3.amazonaws.com/awsmedia/architecturecenter/AWS_ac_ra_web
_01.pdf
● https://s3.amazonaws.com/awslambda-serverless-web-refarch/RefArch_B
logApp_Serverless.png
● https://d0.awsstatic.com/diagrams/product-page-diagrams/Diagrams_gre
engrass-core.png
● http://searchdatacenter.techtarget.com/definition/edge-computing
Homework
● Virtual Machines
○ Read about them and see how they differ from your typical computers
● Check out Azure and Google Cloud Platform
○ See what services they have to offer
○ See how they differ from AWS
● Create your own EC2 Instance
Create an AWS Account
Requirements
● What you need to sign-up for an account:
○ Email
○ Debit/Credit Card
■ You will not charged as long as you stay within usage limits
○ Phone Number
AWS Free Tier
● What you get for free for 12 months:
○ EC2 - 750 Hours
○ RDS - 750 Hours
○ S3 - 5 GB
○ And more!!
● What you get for free forever:
○ CodeCommit - 5 active users/month
○ Lambda - 1 Million requests/month
○ DynamoDB - 25 GB
○ Even more!!
AWS Educate
● What is it?
○ Amazon’s initiative to provide students with resources to learn how to use the cloud
○ Provides credits to use on AWS services not available to free tier
● Who is eligible?
○ Educators
○ Academic Researchers
○ Students
● What you get:
○ Up to $150 in credits
○ Training Course
○ More!!
AWS Educate Starter Account
● Types of Educate Accounts
○ AWS Account
○ AWS Educate Starter Account
● AWS Account
○ More credits
○ Requires credit card
○ Account persists after credits run out
● AWS Educate Starter Accounts
○ Less credits
○ No credit card required
○ Account closes after credits run out
Resources
● AWS resources provided in the readings!
○ AWS Free Tier
○ AWS Educate
AWS Permissions: IAM Identities
Identity and Access Management (IAM) Identities
● What are they?
○ Ways to organize permissions for diferrent resources
● What’s the point?
○ So that only admins have full access while new users have restricted access
○ Ex: Preventing a new hire from nuking your entire database
Identity and Access Management (IAM) Identities
● IAM Users
○ Name and Password
○ Access keys for API or CLI
○ Ex: Bob Seds
● IAM Groups
○ Collection of users
○ Everyone in group inherits the policies of that group
○ Managing policies related to common groups e.g. Admins, Database Team
● IAM Roles
○ Similar to users but can be assumed by any user
○ No credentials
○ Ex: Database Manager
IAM Users vs IAM Roles
● IAM Users
○ Only person working on an account
○ Multiple users
○ Want to use the command-line interface
● IAM Roles
○ Applications
○ Temporary acess
AWS Account Root User
● Spawned when you first create your account
● Full unrestricted access
● Not recommended for everyday use
● Instead, create an IAM user for yourself
Takeaways
● Separate levels of access to resources
● Create an IAM user for yourself
● Don’t use the root user for everyday purposes
● Keep your credentials secure
AWS Permissions: IAM Policies
Identity and Access Management (IAM) Policy
● What is it?
○ Document that defines permissions for certain users, groups, roles, or resources
● Components (Straight from AWS):
○ Effect – whether the policy allows or denies access
○ Action – the list of actions that are allowed or denied by the policy
○ Resource – the list of resources on which the actions can occur
○ Condition (Optional) – the circumstances under which the policy grants permission
Identity and Access Management (IAM) Policy
● What they look like?
○ Written in JSON
Identity and Access Management (IAM) Policy
● How to use them:
○ Attach to a IAM User or Group
○ Attach to a Resource
■ Needs to specify who is affected
■ Specified in principal field
Takeaways
● Assign IAM policies to specify level of access
● Great to limit the possibility of catastrophic accidents
AWS CLI and the SDK's
AWS Command Line Interface (CLI)
● Tool that provides commands to interact with AWS
● Same functionality that’s available on AWS Management Console
● Available on:
○ Linux terminal
○ MacOS terminal
○ Windows PowerShell
Takeaways
Homework
● Create an AWS account
● Create an IAM User with Admin Access
● Install the AWS CLI and configure it
REST API & JSON
REST

REST API

JSON

Amazon AWS
JSON
● Data Structure like XML (SOAP)

● JavaScript Object Notation

JSON
● Objects = { , , } - A collection of name/value pairs

● Arrays = [ , , ] - An ordered list of values

● Object = { “firstname:Jane”, “lastname”:”Smith”}

● Array = [ name1, name2, name3]

name1 = { “firstname:Jane”, “lastname”:”Smith”}

name2 = { “firstname:Kim”, “lastname”:”Estes”}
XML
<person>
<firstName>John</firstName>
<lastName>Smith</lastName>
<age>25</age>
<address>
<streetAddress>21 2nd Street</streetAddress>
<city>New York</city>
<state>NY</state>
<postalCode>10021</postalCode>
</address>
<phoneNumbers>
<phoneNumber type="home">212 555-1234</phoneNumber>
</phoneNumbers>
</person>
JSON
{
"firstName": "John", "lastName": "Smith",
"age": 25,
"address": {
"streetAddress": "21 2nd Street",
"city": "New York",
"state": "NY",
"postalCode": "10021"
},
"phoneNumber": [
{
"type": "home",
"number": "212 555-1234"
}
]
}
JSON
person = { "firstName":"John" , "lastName":"Doe" }

phoneNumber": [
{
"type": "home",
"number": "212 555-1234"
},
{
"type": "fax",
"number": "646 555-4567"
}
]
JSON
student = { "Firstname" : "John", "Lastname" : "Starks", "sid" : "9343934"
}

class = [ { "Firstname" : "John", "Lastname" : "Starks", "sid" : "9343934" },

{ "Firstname" : "Jane", "Lastname" : "Doe", "sid" : "9343934" }

]

= [ student1, student2, …., ]

JSON
● www.json.org

● https://www.sitepoint.com/10-example-json-files/
RESTful API
● Representational state transfer

● RESTful Web services

● http://bit.ly/2gAWSK7
Tesla Motors REST Api
● http://docs.timdorr.apiary.io/
● http://bit.ly/2yI1p85

REST API

Tesla Servers
REST

REST API

Amazon AWS
JSON Framework
● https://github.com/miloyip/nativejson-benchmark C++

● https://github.com/google/gson - Java

● http://bsonspec.org/ - Binary JSON

MQTT Protocol
Overview
● Publisher Subscriber vs Broker
● MQTT
Publisher Subscriber Models
Pub/Sub Message Broker
MQTT vs HTTP
● MQTT exists over TCP/IP, like
http
● Can specify QoS
○ QoS=0 => no confirmation
○ QoS=1 => get packet recieved
confirmation
○ QoS =2 => handshake, exactly one
confirmation (http)
● clientId needs to be unique: use
mac address
● Do not send user/pass in
plaintext. Set up SSL/TLS asap
MQTT Process
Resources and Takeaways
● https://www.hivemq.com
● https://stackoverflow.com/questions/32538535/node-and-mqtt-do-somet
hing-on-message
AWS IoT SDK
Overview
● AWS IoT
● AWS IoT SDK
IoT Devices
IoT Architecture
IoT Architecture
AWS IoT SDK
● https://github.com/aws/aws-iot-device-sdk-python
Resources and Takeaways
● https://d0.awsstatic.com/IoT/assets/awsiot_how_it_works_diagram.png
● https://media.amazonwebservices.com/blog/2017/IoTEnterprise-05-IoTBu
ttonArchitecture-sm.png
Homework
● Read about REST API’s
● Write some JSON
● Write some XML
● Play with the GroupKt Countries API
● Read about MQTT
● Read about AWS IoT
Homework
● Read EC2 documentation
● Read about Apache
● Write some MySQL code
● Write some PHP code
● Read RDS documentation
● Read S3 documentation
● Read Code Commit documentation
● Play around with using all of the above
Rekognition
Overview
● Describe Rekognition
● Upload image via S3
● Detect labels using AWS CLI
● Detect labels using AWS SDK for Python
Polly
Overview
● Describe Polly
● Use Polly through Python:
○ Different phrases
○ Different speakers
Lex
Overview
● Describe Lex
● Create a Lex chatbot on the console
● Interact with that bot through Python
Lex
● Allows for conversational interfaces
○ Essentially chat bots that help you order, schedule, etc.
● Can understand audio and text
● Responds with pre-set messages
● Lex is what powers Amazon Alexa
Important Notes
● Currently only available in certain regions:
○ N. Virginia - us-east-1
○ Ireland - eu-west-1
● Simply change your region in the console and SDK
○ On console, select your current region and change it to US
East (N. Virginia)
○ On SDK, run aws configure and enter us-east-1 for region
● Enable your IAM user to have:
○ AmazonLexFullAccess
GPU’s for Machine Learning
Reasoning
● Why use GPU’s?
○ Much faster than using CPU’s
● Why use AWS instead of a home computer?
○ Training machine learning models is very long
Overview
● Create a GPU instance with Deep Learning libraries pre-installed.
● Clone the Tensorflow repository
● Train a simple model for recognizing digits.
Homework
● Read about Rekognition
● Read about Lex
● Read about Polly
● Read about Machine Learning AMI’s
● Play with all the above