AWS Cloud Practitioner Vocab

Traditional Data Centers

Cons:
- Large investment
- difficult demand forecasting
- slow to deploy
- expensive maintenance
- own all security burden

Cloud Computing

on-demand delivery of comp power, db, apps, and IT through internet
Pros:
- variable expense not capital
- economies of scale

Elasticity

ability to acquire and release resources efficiently

IaaS

The full application and dependencies on the cloud

SaaS

cloud services like email

PaaS

Wordpress, Elastic beanstalks

Deployment Models

Public, Private, Hybrid are types of...

Regions, Availability Zones, Edge Locations

The 3 elements of the AWS Global Infrastructure

Regions

22 so far, cluster of data centers

Availability Zones

At least 2 per region; has at least 1 data center in it; creates no single point of failure

country-region-region number followed by availability zone

AZ naming scheme

Edge Locations

AWS endpoints; consists of CloudFront, Content Delivery Network (CDN), and Route 53

Capital Expenditure

upfront investments to obtain fixed asset

operating expenditure

day to day expenses

AWS Cost Explorer

- A tool that enables you to visualize, understand, and manage your AWS costs and usage over time
- breakdown by service or cost tag
- predictions for next three months

AWS Budgets

- gives you the ability to set custom budgets that alert you when your costs or usage exceed (or are forecasted to exceed) your budgeted amount
- play and track usage by service

AWS TCO Calculator

- Total Cost Ownership
- calculator for savings from data center to cloud

Resource tags

meta data to group projects when calculating costs

AWS Organizations

- An account management service that enables you to consolidate multiple AWS accounts
- multiple accounts under master for Consolidated Billing

AWS Pricing Calculator

- lets you explore AWS services, and create an estimate for the cost of your use cases on AWS.
- deprecated version known as Simple Monthly Calculator

AWS Support

- has 2 services: AWS Personal Health Dashboard & AWS Trusted Advisor
- resource suppport

AWS Personal Health Dashboard

- service that provides alerts and remediation for impacting events

AWS Trusted Advisor

- service that compares usage against best practice; 7 core checks
- provides personalized recommendations

Trusted Advisor checks

5 checks:
- cost optimization
- performance
- security
- fault tolerance
- service limits

Basic, Developer, Business, Enterprise

4 tiers of support

Support differences

- communication method
- response time
- cost
- type of guidance offered

Basic Support

- support plan with 7 trusted advisor checks, personal health dashboard, docs/support forms
- NO support engineers
- Free

Developer Support

- support plan that includes all in basic support
- 1 primary contact
- business hour email access to support engineers
- starts at $29/mo

Business Support

- support plan that includes all in basic and developer
- Full Trusted Advisor Checks
- 24/7 phone, email, chat access
- unlimited contacts
- third part software support
- starts at $100/mo

Enterprise Support

- support plan that includes all in basic, developer, and business
- designed Technical Account Manager (TAM)
- concierge support team
- starts at $15k/mo

Response Times for Developer Plan

- General Guidance: 24 business hours
- System Impaired: 12 business hours
- Prod System Impaired: NA
- Prod System Down: NA
- Business Critical System Down: NA

Response Times for Business Plan

- General Guidance: 24 total hours
- System Impaired: 12 total hours
- Prod System Impaired: 4 hours
- Prod System Down: 1 hour
- Business Critical System Down: NA

Response Times for Enterprise Plan

- General Guidance: 24 total hours
- System Impaired: 12 total hours
- Prod System Impaired: 4 hours
- Prod System Down: 1 hour
- Business Critical System Down: 15 min

AWS Quick Starts

step by step deployment for common platform

AWS Partner Network Consulting Partners

third party connection tool

AWS Professional Services

- a global team of experts that can help you with your desired business outcomes for AWS
- help directly with AWS resources

Console, CLI, SDK

3 ways to interact with AWS

Console

interaction method good for testing out a service

CLI and SDK

interaction method(s) good for automation

SDK

interaction method for app integration

Root and IAM

2 types of AWS users

Identity Access Management

IAM

Compute Services

service to use cloud based VMs; web, db, data processing

EC2

- IaaS service
- Elastic Compute Cloud

Hosting web server, batch processing, API server, remote desktop

4 use cases for EC2

Instance type

- defines processor, memory, storage
- cannot change without downtime

General purpose, compute, memory, or storage optimized, accelerated computing

3 categories of instance types

Root Device Types

- instance store
- elastic block store (EBS)

Instance Store

ephemeral storage; physically attached to host

Amazon Machine Image

AMI

Amazon Machine Image

- provides the information required to launch an instance
- template for EC2 instance like config, OS, data
- can be shared across accounts
- can be custom; commercial ones in AWS marketplace

on-demand, reserved, savings plan, spot, dedicated

- 5 types of EC2 purchases
- default is on-demand

Reserved

- EC2 purchase type for consistent instances
- discount from on-demand when committing to a specific period of time
- capacity reservation for specific instance type

Savings Plan

- purchase types not limited to EC2
- is used for EC2, Fargate, Lambda
- no reserved capacity
- up to 72% savings compared to on-demand
- 1 or 3 year terms

Spot

- purchase type mostly used for batch processing
- up to 90% savings compared to on-demand
- like stock market for instance
- instances launch with highest bidder; otherwise, kicked off
- 2 minutes notify

Dedicated Host

- most expensive purchase type
- dedicated physical server
- good if you have per server license

Elastic Beanstalk

- service that automates deployment & scaling process on EC2
- leverages existing services; only pay for other services
- handles provisioning, load balancing, scaling, and monitoring
Supports Java, .NET, PHP, Node.js, Python, Ruby, Go, Docker
- EC2 customization feature
- to deploy apps with minimal knowledge of other services
- reduces app maintenance

AWS Lambda

- compute service where you are charged on execution time
- enables event-driven workflows
- primary service for serverless architecture
Advantages: reduced maintenance, fault tolerance, scales based on demand

Amazon Route 53, Amazon VPC, AWS Direct Connect, Amazon API Gateway, Amazon CloudFront, Elastic Load Balancing

6 types of Content and Network Delivery Services

Amazon VPC

- isolated section of the AWS Cloud
- supports IPv4 and IPv6; can configure network settings
- supports public and private subnets
- can utilize Network Address Translation (NAT) for private subnets
- enables connections to data center
- can connect to other VPCs
- supports private connection to other AWS Services

AWS Direct Connect

- service for dedicated network connection from data center to AWS

Amazon Route 53

- highly available and scalable DNS web service
- leverage edge locations
- global service; highly available and rare downtimes
- global resource routing
- can use server fail backups

Elastic Load Balancing

- distributes traffic across multiple targets
- integrates with EC2, ECS (Docker Service), Lambda
- supports one of more AZs in a region

Application Load Balancer, Network Load Balancer, Classic Load Balancer

3 types of load balancers; default is classic (known as ELB)

vertical and horizontal

2 EC2 scaling types

vertical scaling

scaling to better instance types

horizontal scaling

adding more of the same instance type

Amazon CloudFront

- CDN
- leverages edge locations
- for static and dynamic content
- security features: AWS Shield for DDoS
- AWS Web Application Firewall (WAF)

Amazon API Gateway

- fully managed API management service
- can create and manage APIs
- Directly integrates with other AWS services
- API call monitoring and metrics
- Supports VPC and on-premise private applications

AWS Global Accelerator

- networking service that improves user performance up to 60%
- utilizes IP addresses instead of DNS like CloudFront
- once traffic reaches edge locations, traffic is routed through AWS network instead of public internet
- can route to AWS resources like load balancers and EC2 instances

Global Accelerator

performance improvements from this service by:
- minimizing distance between user and endpoint by using edge locations
- optimizing traffic using AWS network
- improvement of first byte latency, jitter, throughput
- superior fault tolerance by not relying on DNS resolution

Global Accelerator

use cases for this network service:
- using non-http protocols
- requiring static IP
- instant failover

Amazon S3, Amazon S3 Glacier, Amazon Elastic Block Store, Amazon Elastic File System, AWS Snowball, AWS Snowmobile

6 types of file storage services

Amazon S3

- storage service used to store files as objects in buckets
- storage classes for different use cases
- stores data across multiple AZs
- URL access for files
- configurable rules for data lifecycle
- can serve as static web host

standard, intelligent-tiering, standard-IA, One Zone-IA

4 types of S3 non-archival classes

Standard

- S3 default non-archival storage class
- used for frequently accessed data

Intelligent-Tiering

- S3 non-archival storage class
- moves data to storage class based on usage based on frequency access

Standard-IA

- S3 non-archival storage class
- for infrequently accessed data
- spread across multiple AZs

One Zone-IA

- S3 non-archival storage class
- for infrequently accessed data
- only in one AZ

Lifecycle Policies

For S3:
- bucket objects can transition or expire based on criteria
- transition based on time not usage
- expiration on age
- policies can account for versions of an object

Transfer Acceleration

optimized data upload using edge location as a part of CloudFront

Glacier & Glacier Deep Archive

- 2 services for archiving data within S3 as separate storage classes
- offers configurable retrieval times
- can send files directly or through lifecycle rules

S3 Glacier

- service for archival data
- 90 day min storage duration change
- retrieved in minutes or hours
- retrieval fee per GB
- 5x LESS expensive than S3 standard storage class

S3 Glacier Deep Archive

- for archival data
- 180 day min storage duration change
- retrieved in hours
- retrieval free per GB
- 23x LESS expensive than S3 standard storage class

Elastic Block Store

- persistent storage; separate from host
- persistent block storage for use with a single EC2 instance
- scales to support PB of data & multiple volume types
- redundancy within AZ
- allows snapshots of data
- offers encryptions

EBS volume types

1. General purpose SSD
2. Provisioned IOPS SSD; high performance for low latency apps
3. Throughput optimized HDD; frequently accessed data
4. Cold HDD; less frequently accessed workloads

Elastic File System (EFS)

- fully managed Network File System (NFS)
- Designed for Linux workloads
- Supports PB of data
- Stores data across multiple AZs
- provides configurable lifecycles rules

Standard, Infrequent Access

2 EBS Storage Classes

Amazon FSx for Windows File Server

- fully managed native Windows file system
- Utilizes SSDs for low latency
- Native Windows features like:
- SMB (Server Message Block) support
- Active Directory Integration
- Windows NTFS

AWS Snowball

- large scale data transfer
- PB scale transfers
- physical device delivered by AWS to office location
- connect Snowball to your network and upload data
- Device is returned by local carrier to AWS
- AWS receives device and loads data into S3

AWS Snowmobile

- large scale data transfer
- EB scale transfers
- shipping container delivered to location
- AWS sets up connection to network
- loads data onto snowmobile
- AWS loads data into S3 when the container is at AWS location
- can do multiple trips

Amazon RDS, Amazon Aurora, Amazon DynamoDB, Amazon Redshift, Elasticache, AWS Database Migration Service

6 database and utility services

Amazon RDS

- fully managed service for relational databases
- handles provisioning, patching, backups, and recovery
- can deploy across multiple AZs
- supports read replicas for some platforms
- launches into VPC
- provides both GP SSD and provisioned IOPS SSDs

MySQL, PostgresSQL, MariaDB, Oracle, SQL Server, Amazon Aurora

6 Amazon RDS supported platforms

Amazon DMS (Data Migration Service)

- service to move data into AWS from existing data
- supports one time and continual migration
- supports popular commercial and open source databases
- only pay for computation in migration process

Amazon DynamoDB

- fully managed NoSQL database service
- provides key-value and document DB
- low latency at any scale
- supports automated scaling based on config
- offers in-memory cache with DynamoDB Accelerator (DAX)
- Handles 20 mil reqs/sec and 10 tril reqs/day

Amazon DynamoDB

- NoSQL database service use cases:
- scaling without excessive maintenance
- serverless applications
- implementations that require low latency
- data models without BLOB (Binary Large Object) storage

Amazon ElastiCache

- fully managed in-memory data store
- supports both Memcached and Redis
- low latency
- enables scaling and replicas to meet demand
- handles common use cases:
- DB layer caching
- session storage

Amazon Redshift

- scalable data warehouse service
- supports PB scale warehousing
- high performance disks and columnar storage
- offers full encryption for content
- provides isolation with a VPC
- enables querying of EB of data in S3 using Redshift Spectrum

Amazon Aurora

- relational DB engine compatible with MySQL and PostgreSQL