Page 1
© 2014 Amazon.com, Inc. and its affiliates. All rights reserved. May not be copied, modified, or distributed in whole or in part without the express consent of Amazon.com, Inc.
Introduction to Amazon EC2
Stephen Elliott, Amazon Web Services
March 26th 2014
Page 2
What we’ll cover:
• A brief look back
• Amazon EC2– Key functionality
– Companion services
• Demo
Page 3
What is Amazon EC2..?
Page 4
A Brief look back…
Amazon EC2: Seven Years Young
Page 5
Amazon EC2 – Seven years ago
• Single instance family and size– m1.small (1 vCPU, 1.7 GiB RAM, 160 GB storage)
• Linux only
• On Demand pricing only
• No AWS Management Console
• No EBS
• No Elastic IPs
• No VPC
• No Auto Scaling
• No Elastic Load BalancingAmazon EC2
Page 6
Since then• Many new families and sizes
– M1, M2, C1, C3, M3, CC1, CC2, CG1, G2, CR1, HS1, HI1, T1, G2, I2
• Windows (and Enterprise Linux)
• Reserved Instances and Spot Instances
• AWS Management Console
• Elastic Block Store
• Elastic IP Addresses
• Virtual Private Cloud (VPC)
• Auto Scaling
• Elastic Load Balancing
• Performance, security, manageability, and scalability improvements
Page 7
AWS Global Infrastructure
10 Regions
25 Availability Zones
51 Edge locations
Page 8
Amazon EC2
• Resizable compute instances in the cloud
• Provision 1 or many instances
• Pay for what you use; no minimum commitment
• Familiar operating systems, with cloud benefits
Page 9
Amazon EC2 Instances - Families
Compute-Optimized
Storage-Optimized
Micro Instances
General Purpose
GPU Instances
Memory-Optimized
Page 10
C1/CC2/C3
HI1/I2/HS1
T1
M1/M3
G2
M2/CR1
Amazon EC2 Instances - Generations
Page 11
Amazon EC2 Instances – Types*
c3.large
i2.xlarge
t1.micro
m3.medium
g2.2xlarge
m2.xlarge
*Note: Only instance type is shown from each family/generation, as an example
Page 12
Recent Instance Highlights
• HI1– 120,000 random read IOPS on SSDs
• I2 - Next generation I/O optimized (350,000+ random read IOPS)
• M3 – Next generation general purpose
• HS1– 48 TB of storage per instance
• CR1– 244 GiB of RAM &Intel Xeon E5 2670 (Sandy Bridge) CPUs
• G2 – 1 NVIDIA GK104 GPU (Kepler)
• C3 - 2.8 GHz Intel Xeon E5-2680v2 (Ivy Bridge) CPUs (SSD)
Page 13
A choice of block storage options
• Locally attached, “instance storage”– Type/amount differs by instance
• Amazon EBS Standard Volumes
• Amazon EBS Provisioned IOPS Volumes
Page 15
Using AWS, Dow Jones saved $40,000;
improves time-to-market by 30%
• Dow Jones & Company provides news and
business information for the global investment
market.
• Dow Jones needed to replace its Asian data
center with a flexible, cost-effective alternative
that allowed its staff to focus on building
products
• Moving its data center to AWS saved $40,000 in
hardware costs and enabled Dow Jones staff to
focus on creating revenue-producing
applications for the website
Page 16
• One of the world's leading online travel
companies, providing leisure and business
travel to customers worldwide
• Expedia needed a global infrastructure to
support its commitment to create a great
experience for customers worldwide
• Using AWS, Expedia created a global
infrastructure for its critical applications while
reducing network latency to under 50
milliseconds and enabling a 230% CPU
consumption efficiency for data processing
Expedia uses AWS to run critical, high volume applications globally
Page 17
Some Amazon EC2 Benefits
• Elastic
• Completely Controlled
• Flexible
• Reliable
• Secure
• Inexpensive
Page 19
Amazon EC2 is designed to help you:
• Simplify your operations
• Scale as required
• Improve resiliency
• Run applications securely
• Run any application
• Reduce your costs
Page 20
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Serv
er
Lo
ad
Hour of day
Page 21
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Serv
er
Lo
ad
Hour of day
Capacity of 1 Server
Page 22
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Serv
er
Lo
ad
Hour of day
Capacity of 1 Server
Traditional capacity required
Page 23
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Serv
er
Lo
ad
Hour of day
Capacity of 1 Server
Traditional capacity required
1 Server for 8 hours
Page 24
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Serv
er
Lo
ad
Hour of day
Capacity of 1 Server
Traditional capacity required
1 Server for 8 hours 1 Server for 8 hours
Page 25
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Serv
er
Lo
ad
Hour of day
Capacity of 1 Server
Traditional capacity required
1 Server for 8 hours 1 Server for 8 hours
1 Server for 8 hours
Page 26
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Serv
er
Lo
ad
Hour of day
Capacity of 1 Server
Traditional capacity required
1 Server for 8 hours 1 Server for 8 hours
1 Server for 8 hours
1 Server for 8 hours
Page 27
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Serv
er
Lo
ad
Hour of day
Capacity of 1 Server
Traditional capacity required
1/3rd
Saving
Page 28
0
1
2
3
4
5
6
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Insta
nce C
ou
nt
Day of Month
Page 29
0
1
2
3
4
5
6
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Insta
nce C
ou
nt
Day of Month
Monthly
predictable
peak
processing
Page 30
0
1
2
3
4
5
6
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Insta
nce C
ou
nt
Day of Month
Traditional capacity required
Page 31
0
1
2
3
4
5
6
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Insta
nce C
ou
nt
Day of Month
Elastic Capacity
Traditional capacity required
Page 32
0
1
2
3
4
5
6
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Insta
nce C
ou
nt
Day of Month
75% Savings
Traditional capacity required
Elastic Capacity
Page 33
Completely Controlled
Page 34
Completely Controlled
• You have control of your instances
• Log on as root (Linux) / Administrator (Windows)
• Install the software you need
• Make the configuration changes you like
• Create an AMI (Amazon Machine Image)
• Start/Stop and control via console or APIs
Page 37
Multiple Instance Types
• Choose the instance type that suits you
• Change the instance type when you want to
• Attach as much or as little storage as you need
• Choose your operating system
• Choose a pre-configured image (AMI)
Page 39
Build Reliable Architectures
• Easily build highly available applications
• AWS Elastic Load Balancing distributes load
• Auto Scaling helps ensure availability and scale
• Use multiple Availability Zones (AZs)
Page 40
Example 3-tier Web App architecture
Page 42
Security
Certifications:
• Secured premises
• Secured storage
• Secured keys
• Secured communications
• Secured network access
• Secured users
• Secured logins
• Full control to add security
Page 43
Amazon Virtual Private Cloud (VPC) - Default
• Provision a logically isolated section of the AWS cloud
• Control your virtual networking environment– Subnets
– Route Tables
– Security Groups
– Network ACLs
• Connect to your on-premises network via hw VPN
• Control if and how your instances access the Internet
Page 44
Availability Zone 1a Availability Zone 1b
Internet
10.0.0.5
10.0.0.6
10.0.3.17
10.0.3.5
10.0.1.5
10.0.1.25
10.0.1.8
10.0.1.6
VPC Subnet
VPC Subnet
VPC Subnet
Virtual Private Gateway
Customer Gateway
VPN Connection
Internet Gateway
Customer Data Center
Page 46
Unix/Linux instances start at $0.02/hour
Pay as you go for compute power
Low cost and flexibility
Pay only for what you use, no up-front commitments or long-term contracts
Use Cases:
Applications with short term, spiky, or unpredictable workloads;
Application development or testing
On-demand instances
On-Demand instances
Page 47
Unix/Linux instances start at $0.02/hour
Pay as you go for compute power
Low cost and flexibility
Pay only for what you use, no up-front commitments or long-term contracts
Use Cases:
Applications with short term, spiky, or unpredictable workloads;
Application development or testing
On-demand instances
1- or 3-year terms
Pay low up-front fee, receive significant hourly discount
Low Cost / Predictability
Helps ensure compute capacity is available when needed
Use Cases:
Applications with steady state or predictable usage
Applications that require reserved capacity, including disaster recovery
Reserved instances
Reserved instances
Page 48
Unix/Linux instances start at $0.02/hour
Pay as you go for compute power
Low cost and flexibility
Pay only for what you use, no up-front commitments or long-term contracts
Use Cases:
Applications with short term, spiky, or unpredictable workloads;
Application development or testing
On-demand instances
1- or 3-year terms
Pay low up-front fee, receive significant hourly discount
Low Cost / Predictability
Helps ensure compute capacity is available when needed
Use Cases:
Applications with steady state or predictable usage
Applications that require reserved capacity, including disaster recovery
Reserved instances
Reserved instances
Up to 58% Savings
Heavy utilization RI
Page 49
Unix/Linux instances start at $0.02/hour
Pay as you go for compute power
Low cost and flexibility
Pay only for what you use, no up-front commitments or long-term contracts
Use Cases:
Applications with short term, spiky, or unpredictable workloads;
Application development or testing
On-demand instances
1- or 3-year terms
Pay low up-front fee, receive significant hourly discount
Low Cost / Predictability
Helps ensure compute capacity is available when needed
Use Cases:
Applications with steady state or predictable usage
Applications that require reserved capacity, including disaster recovery
Reserved instances
Reserved instances
> 80% utilizationLower costs up to 58%
Use Cases: Databases, Large Scale HPC, Always-on infrastructure, Baseline
Heavy utilization RI
Page 50
Unix/Linux instances start at $0.02/hour
Pay as you go for compute power
Low cost and flexibility
Pay only for what you use, no up-front commitments or long-term contracts
Use Cases:
Applications with short term, spiky, or unpredictable workloads;
Application development or testing
On-demand instances
1- or 3-year terms
Pay low up-front fee, receive significant hourly discount
Low Cost / Predictability
Helps ensure compute capacity is available when needed
Use Cases:
Applications with steady state or predictable usage
Applications that require reserved capacity, including disaster recovery
Reserved instances
Reserved instances
> 80% utilizationLower costs up to 58%
Use Cases: Databases, Large Scale HPC, Always-on infrastructure, Baseline
Heavy utilization RI
Up to 49%
Savings
Medium utilization RI
Page 51
Unix/Linux instances start at $0.02/hour
Pay as you go for compute power
Low cost and flexibility
Pay only for what you use, no up-front commitments or long-term contracts
Use Cases:
Applications with short term, spiky, or unpredictable workloads;
Application development or testing
On-demand instances
1- or 3-year terms
Pay low up-front fee, receive significant hourly discount
Low Cost / Predictability
Helps ensure compute capacity is available when needed
Use Cases:
Applications with steady state or predictable usage
Applications that require reserved capacity, including disaster recovery
Reserved instances
Reserved instances
> 80% utilizationLower costs up to 58%
Use Cases: Databases, Large Scale HPC, Always-on infrastructure, Baseline
Heavy utilization RI
41-79% utilizationLower costs up to 49%
Use Cases: Web applications, many heavy
processing tasks, running much of the time
Medium utilization RI
Page 52
Unix/Linux instances start at $0.02/hour
Pay as you go for compute power
Low cost and flexibility
Pay only for what you use, no up-front commitments or long-term contracts
Use Cases:
Applications with short term, spiky, or unpredictable workloads;
Application development or testing
On-demand instances
1- or 3-year terms
Pay low up-front fee, receive significant hourly discount
Low Cost / Predictability
Helps ensure compute capacity is available when needed
Use Cases:
Applications with steady state or predictable usage
Applications that require reserved capacity, including disaster recovery
Reserved instances
Reserved instances
> 80% utilizationLower costs up to 58%
Use Cases: Databases, Large Scale HPC, Always-on infrastructure, Baseline
Heavy utilization RI
41-79% utilizationLower costs up to 49%
Use Cases: Web applications, many heavy
processing tasks, running much of the time
Medium utilization RI
15-40% utilizationLower costs up to 34%
Use Cases: Disaster Recovery, Weekly / Monthly reporting, Elastic Map Reduce
Light utilization RI
Page 53
Improving Reserved Instance Flexibility
Page 54
Best RI for Utilization
$-
$2,000
$4,000
$6,000
$8,000
$10,000
$12,000
$14,000
$16,000
$18,000
Heavy
Medium
Light
O-Demand
Page 55
Best RI for Utilization
$-
$2,000
$4,000
$6,000
$8,000
$10,000
$12,000
$14,000
$16,000
$18,000
Heavy
Medium
Light
O-Demand
Page 56
0
2
4
6
8
10
12
14
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
On Demand
Light Utilization RI
Medium Utilization RI
Heavy utilization RI
Optimizing costs with RIs
Page 57
Unix/Linux instances start at $0.02/hour
Pay as you go for compute power
Low cost and flexibility
Pay only for what you use, no up-front commitments or long-term contracts
Use Cases:
Applications with short term, spiky, or unpredictable workloads;
Application development or testing
On-demand instances
1- or 3-year terms
Pay low up-front fee, receive significant hourly discount
Low Cost / Predictability
Helps ensure compute capacity is available when needed
Use Cases:
Applications with steady state or predictable usage
Applications that require reserved capacity, including disaster recovery
Reserved instances
Bid on unused EC2 capacity
Spot Price based on supply/demand, determined automatically
Cost / Large Scale, dynamic workload handling
Use Cases:
Applications with flexible start and end times
Applications only feasible at very low compute prices
Spot instances
Spot instances
Page 58
We like to save you money…
Page 59
EC2 Usage Reports
• Two new interactive reports:– Instance Usage Report
– Reserved Instance Utilization Report
• Easy to access in the Management Console
Page 60
EC2 Usage Reports
Page 61
EC2 Usage Reports
Page 62
Easy to get started..!
Page 66
AWS Marketplace – in the console
Browse, search, discover
and launch thousand of AWS
Marketplace AMIs directly
from within the Amazon EC2
console.
Page 67
AWS Marketplace
• AWS Online Software Store
• Find, research and buy software
• Simple pricing, aligns with the utility model
• 1-Click launch products - run in minutes
• Over 1300 products listed in 25 categories
• Free trials and Enterprise offerings– Move seamlessly from PoC to production
Page 68
Benefits for You
• 1-Click launch, no
procurement of
hardware and software
launches pre-configured
for AWS
• Minutes for deployment,
instead of weeks or
months
• Vetted, tested, secure
and approved products
• Use your AWS account
to purchase
• Supports Invoice Billing
• See charges for
software and AWS
services separately
• Clear pricing for all products, hourly and monthly options, different instance size and region choices
• Tools to track your purchases and control usage
• Ability to scale software up or down on-demand
Find and Deploy Fast One Bill Simple Purchasing
http://aws.amazon.com/marketplace
Page 69
University of Liège is a Belgian multi-disciplinary research lab studying the brain’s function and structure
Needed a solution that would quickly and reliably conduct image processing for Parkinson’s research
Found the NITRC computational environment on AWS Marketplace, reducing time required to process neuroimaging data by 85%
Using NITRC on AWS Marketplace, University of Liège
Accelerates Parkinson’s Disease Research
Page 70
Next Steps
• Sign-up for an AWS account!
• Take advantage of the free tier– aws.amazon.com/free
• Learn more: aws.amazon.com/ec2
• Join us at the booth
Page 71
© 2014 Amazon.com, Inc. and its affiliates. All rights reserved. May not be copied, modified, or distributed in whole or in part without the express consent of Amazon.com, Inc.
Introduction to Amazon EC2
Stephen Elliott, Amazon Web Services
March 26th 2014
Thank you!