Aws Storage Options

Storage Options in the
AWS Cloud
October 2013
© 2016, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Notices
This document is provided for informational purposes only. It represents AWS’s
current product offerings and practices as of the date of issue of this document,
which are subject to change without notice. Customers are responsible for
making their own independent assessment of the information in this document
and any use of AWS’s products or services, each of which is provided “as is”
without warranty of any kind, whether express or implied. This document does
not create any warranties, representations, contractual commitments,
conditions or assurances from AWS, its affiliates, suppliers or licensors. The
responsibilities and liabilities of AWS to its customers are controlled by AWS
agreements, and this document is not part of, nor does it modify, any agreement
between AWS and its customers.
Contents
Introduction 1
Traditional vs. Cloud-Based Storage Alternatives 1
Amazon Simple Storage Service (Amazon S3) 3
Amazon Glacier 7
Amazon Elastic Block Store (Amazon EBS) Volumes 10
Amazon EC2 Instance Store Volumes 15
AWS Import/Export 20
AWS Storage Gateway 22
Amazon CloudFront 25
Amazon Simple Queue Service (Amazon SQS) 28
Amazon Relational Database Service (Amazon RDS) 31
Amazon DynamoDB 35
Amazon ElastiCache 38
Amazon Redshift 41
Databases on Amazon EC2 44
Contributors 47
Further Reading 47
Cloud Storage Use Cases 47
AWS Storage Services 47
Other Resources 48
Amazon Web Services – Storage Options in the AWS Cloud
Introduction
Amazon Web Services (AWS) is a flexible, cost-effective, easy-to-use cloud
computing platform. This whitepaper helps architects and developers
understand the primary data storage options available in the AWS cloud. We
provide an overview of each storage option, describe ideal usage patterns,
performance, durability and availability, cost model, scalability and elasticity,
interfaces, and anti-patterns.
In a separate companion whitepaper, we present several storage use cases that

show how to use multiple AWS cloud storage options together. You can employ
these use cases as a guide when designing your own storage architecture. You
can see this companion whitepaper at Storage Options in the AWS Cloud: Use
Cases.1
Traditional vs. Cloud-Based Storage Alternatives

Architects of traditional, on-premises IT infrastructures and applications have
numerous potential data storage choices, including the following:
• Memory—In-memory storage, such as file caches, object caches, in-

memory databases, and RAM disks, provide very rapid access to data.
• Message Queues—Temporary durable storage for data sent

asynchronously between computer systems or application components.
• Storage area network (SAN)—Block devices (virtual disk LUNs) on

dedicated SANs often provide the highest level of disk performance and
durability for both business-critical file data and database storage.
• Direct-attached storage (DAS)—Local hard disk drives or arrays
residing in each server provide higher performance than a SAN, but
lower durability for temporary and persistent files, database storage, and
operating system (OS) boot storage than a SAN.
• Network attached storage (NAS)—NAS storage provides a file-level

interface to storage that can be shared across multiple systems. NAS
tends to be slower than either SAN or DAS.
• Databases—Structured data is typically stored in some kind of database,

such as a traditional SQL relational database, a NoSQL non-relational
Page 1 of 50
database, or a data warehouse. The underlying database storage typically

resides on SAN or DAS devices, or in some cases in memory.
• Backup and Archive—Data retained for backup and archival purposes is

typically stored on non-disk media such as tapes or optical media, which
are usually stored off-site in remote secure locations for disaster recovery.
Each of these traditional storage options differs in performance, durability, and

cost, as well as in their interfaces. Architects consider all these factors when
identifying the right storage solution for the task at hand. Notably, most IT
infrastructures and application architectures employ multiple storage
technologies in concert, each of which has been selected to satisfy the needs of a
particular subclass of data storage, or for the storage of data at a particular point
in its lifecycle. These combinations form a hierarchy of data storage tiers.
As we’ll see throughout this whitepaper, AWS offers multiple cloud-based

storage options. Each has a unique combination of performance, durability,
availability, cost, and interface, as well as other characteristics such as scalability
and elasticity. These additional characteristics are critical for web-scale cloud-
based solutions. As with traditional on-premises applications, you can use
multiple cloud storage options together to form a comprehensive data storage
hierarchy.
This whitepaper examines the following AWS cloud storage options:
Amazon S3 Scalable storage in the cloud
Amazon Glacier Low-cost archive storage in the cloud
Persistent block storage volumes for Amazon EC2 virtual

Amazon EBS
machines
Amazon EC2 Instance Temporary block storage volumes for Amazon EC2 virtual
Storage machines
AWS Import/Export Large volume data transfer
AWS Storage Gateway Integrates on-premises IT environments with cloud storage
Amazon CloudFront Global content delivery network (CDN)
Page 2 of 50
Amazon SQS Message queue service
Managed relational database server for MySQL, Oracle, and

Amazon RDS
Microsoft SQL Server
Amazon DynamoDB Fast, predictable, highly-scalable NoSQL data store
Amazon ElastiCache In-memory caching service
Fast, powerful, full-managed, petabyte-scale data warehouse

Amazon Redshift
service
Databases on Amazon
Self-managed database on an Amazon EC2 instance
EC2
For additional comparison categories among the AWS storage collection, see the
AWS Storage Quick Reference.
Amazon Simple Storage Service (Amazon S3)

Amazon Simple Storage Service (Amazon S3) is storage for the Internet.2 It’s a
simple storage service that offers software developers a highly-scalable, reliable,
and low-latency data storage infrastructure at very low costs. Amazon S3
provides a simple web services interface that can be used to store and retrieve
any amount of data, at any time, from within Amazon Elastic Compute Cloud
(Amazon EC2) or from anywhere on the web. You can write, read, and delete
objects containing from 1 byte to 5 terabytes of data each. The number of objects
you can store in an Amazon S3 bucket is virtually unlimited. Amazon S3 is also
highly secure, supporting encryption at rest, and providing multiple
mechanisms to provide fine-grained control of access to Amazon S3 resources.
Amazon S3 is also highly scalable, allowing concurrent read or write access to
Amazon S3 data by many separate clients or application threads. Finally,
Amazon S3 provides data lifecycle management capabilities, allowing users to
define rules to automatically archive Amazon S3 data to Amazon Glacier, or to
delete data at end of life.
Ideal Usage Patterns

One very common use for Amazon S3 is storage and distribution of static web
content and media. This content can be delivered directly from Amazon S3,
since each object in Amazon S3 has a unique HTTP URL address, or Amazon S3
Page 3 of 50
can serve as an origin store for a content delivery network (CDN), such as
Amazon CloudFront. Because of Amazon S3’s elasticity, it works particularly
well for hosting web content with extremely spiky bandwidth demands. Also,
because no storage provisioning is required, Amazon S3 works well for fast
growing websites hosting data intensive, user-generated content, such as video
and photo sharing sites.
Amazon S3 is also frequently used to host entire static websites. Amazon S3

provides a highly-available and highly-scalable solution for websites with only
static content, including HTML files, images, videos, and client-side scripts such
as JavaScript.
Amazon S3 is also commonly used as a data store for computation and large-
scale analytics, such as analyzing financial transactions, clickstream analytics,
and media transcoding. Because of the horizontal scalability of Amazon S3, you
can access your data from multiple computing nodes concurrently without
being constrained by a single connection.
Finally, Amazon S3 is often used as a highly durable, scalable, and secure

solution for backup and archival of critical data, and to provide disaster
recovery solutions for business continuity. Because Amazon S3 stores objects
redundantly on multiple devices across multiple facilities, it provides the
highly-durable storage infrastructure needed for these scenarios. Amazon S3’s
versioning capability is available to protect critical data from inadvertent
deletion.
Performance
Access to Amazon S3 from within Amazon EC2 in the same region is fast.
Amazon S3 is designed so that server-side latencies are insignificant relative to
Internet latencies. Amazon S3 is also built to scale storage, requests, and users to
support a virtually unlimited number of web-scale applications. If you access
Amazon S3 using multiple threads, multiple applications, or multiple clients
concurrently, total Amazon S3 aggregate throughput will typically scale to rates
that far exceed what any single server can generate or consume.
To speed access to relevant data, many developers pair Amazon S3 with a

database, such as Amazon DynamoDB or Amazon RDS. Amazon S3 stores the
actual information, and the database serves as the repository for associated
Page 4 of 50
metadata (e.g., object name, size, keywords, and so on). Metadata in the database
can easily be indexed and queried, making it very efficient to locate an object’s
reference via a database query. This result can then be used to pinpoint and
then retrieve the object itself from Amazon S3.
Durability and Availability

By automatically and synchronously storing your data across both multiple
devices and multiple facilities within your selected geographical region, Amazon
S3 storage provides the highest level of data durability and availability in the
AWS platform. Error correction is built-in, and there are no single points of
failure. Amazon S3 is designed to sustain the concurrent loss of data in two
facilities, making it very well-suited to serve as the primary data storage for
mission-critical data. In fact, Amazon S3 is designed for 99.999999999% (11
nines) durability per object and 99.99% availability over a one-year period. In
addition to its built-in redundancy, Amazon S3 data can also be protected from
application failures and unintended deletions through the use of Amazon S3
versioning. You can also enable Amazon S3 versioning with
Multi-Factor Authentication (MFA) Delete. With this option enabled on a

bucket, two forms of authentication are required to delete a version of an
Amazon S3 object: valid AWS account credentials plus a six-digit code (a single-
use, time-based password) from a physical token device.
For noncritical data that can be reproduced easily if needed, such as transcoded
media or image thumbnails, you can use the Reduced Redundancy Storage
(RRS) option in Amazon S3, which provides a lower level of durability at a lower
storage cost. Objects stored using the RRS option have less redundancy than
objects stored using standard Amazon S3 storage. In either case, your data is
still stored on multiple devices in multiple locations. RRS is designed to provide
99.99% durability per object over a given year. While RRS is less durable than
standard Amazon S3, it is still designed to provide 400 times more durability
than a typical disk drive.
Cost Model
With Amazon S3, you pay only for what you use and there is no minimum fee.
Amazon S3 has three pricing components: storage (per GB per month), data
transfer in or out (per GB per month), and requests (per n thousand requests per
month). For new customers, AWS provides a free usage tier which includes up to
Page 5 of 50
5 GB of Amazon S3 storage.3 Pricing information can be found at Amazon S3

Pricing.4
Scalability and Elasticity

Amazon S3 has been designed to offer a very high level of scalability and
elasticity automatically. Unlike a typical file system that encounters issues when
storing large number of files in a directory, Amazon S3 supports a virtually
unlimited number of files in any bucket. Also, unlike a disk drive that has a
limit on the total amount of data that can be stored before you must partition
the data across drives and/or servers, an Amazon S3 bucket can store a virtually
unlimited number of bytes. You are able to store any number of objects (files)
in a single bucket, and Amazon S3 will automatically manage scaling and
distributing redundant copies of your information to other servers in other
locations in the same region, all using Amazon’s high-performance
infrastructure.
Interfaces
Amazon S3 provides standards-based REST and SOAP web services APIs for
both management and data operations. These APIs allow Amazon S3 objects
(files) to be stored in uniquely-named buckets (top-level folders). Each object
must have a unique object key (file name) that serves as an identifier for the
object within that bucket. While Amazon S3 is a web-based object store rather
than a traditional file system, you can easily emulate a file system hierarchy
(folder1/folder2/file) in Amazon S3 by creating object key names that
correspond to the full path name of each file.
Most developers building applications on Amazon S3 use a higher-level toolkit

or software development kit (SDK) that wraps the underlying REST API. AWS
SDKs are available for Java, .NET, PHP, and Ruby. The integrated AWS
Command Line Interface (AWS CLI) also provides a set of high-level, Linux-like
Amazon S3 file commands for common operations, such as ls, cp, mv, sync, etc.
The AWS CLI for Amazon S3 allows you to perform recursive uploads and
downloads using a single folder-level Amazon S3 command, and supports
parallel transfers. The AWS CLI also provides command-line access to the low-
level Amazon S3 API. Finally, using the AWS Management Console, you can
easily create and manage Amazon S3 buckets, upload and download objects,
and browse the contents of your Amazon S3 buckets using a simple web-based
user interface.
Page 6 of 50
Anti-Patterns
Amazon S3 is optimal for storing numerous classes of information that are
relatively static and benefit from its durability, availability, and elasticity
features. However, in a number of situations Amazon S3 is not the optimal
solution. Amazon S3 has the following anti-patterns:
• File system—Amazon S3 uses a flat namespace and isn’t meant to serve

as a standalone, POSIX-compliant file system. However, by using
delimiters (commonly either the ‘/’ or ‘\’ character) you are able construct
your keys to emulate the hierarchical folder structure of file system
within a given bucket.
• Structured data with query—Amazon S3 doesn’t offer query

capabilities: to retrieve a specific object you need to already know the
bucket name and key. Thus, you can’t use Amazon S3 as a database by
itself. Instead, pair Amazon S3 with a database to index and query
metadata about Amazon S3 buckets and objects.
• Rapidly changing data—Data that must be updated very frequently

might be better served by a storage solution with lower read / write
latencies, such as Amazon EBS volumes, Amazon RDS or other relational
databases, or Amazon DynamoDB.
• Backup and archival storage—Data that requires long-term encrypted

archival storage with infrequent read access may be stored more cost-
effectively in Amazon Glacier.
• Dynamic website hosting—While Amazon S3 is ideal for websites with

only static content, dynamic websites that depend on database
interaction or use server-side scripting should be hosted on Amazon EC2.
Amazon Glacier
Amazon Glacier is an extremely low-cost storage service that provides highly
secure, durable, and flexible storage for data backup and archival.5 With
Amazon Glacier, customers can reliably store their data for as little as $0.01 per
gigabyte per month. Amazon Glacier enables customers to offload the
administrative burdens of operating and scaling storage to AWS, so that they
don’t have to worry about capacity planning, hardware provisioning, data
replication, hardware failure detection and repair, or time-consuming hardware
migrations.
Page 7 of 50
You store data in Amazon Glacier as archives. An archive can represent a single
file or you may choose to combine several files to be uploaded as a single
archive. Retrieving archives from Amazon Glacier requires the initiation of a job.
You organize your archives in vaults. You can control access to your vaults using
the AWS Identity and Access Management (IAM) service.
Amazon Glacier is designed for use with other Amazon Web Services. Amazon
S3 allows you to seamlessly move data between Amazon S3 and Amazon Glacier
using data lifecycle policies. You can also use AWS Import/Export to accelerate
moving large amounts of data into Amazon Glacier using portable storage
devices for transport.

Organizations are using Amazon Glacier to support a number of use cases.
These include archiving offsite enterprise information, media assets, research
and scientific data, digital preservation and magnetic tape replacement.
Performance
Amazon Glacier is a low-cost storage service designed to store data that is
infrequently accessed and long lived. Amazon Glacier jobs typically complete in
3 to 5 hours.

Amazon Glacier is designed to provide average annual durability of
99.999999999% (11 nines) for an archive. The service redundantly stores data
in multiple facilities and on multiple devices within each facility. To increase
durability, Amazon Glacier synchronously stores your data across multiple
facilities before returning SUCCESS on uploading archives. Unlike traditional
systems, which can require laborious data verification and manual repair,
Amazon Glacier performs regular, systematic data integrity checks and is built
to be automatically self-healing.
Cost Model
With Amazon Glacier, you pay only for what you use and there is no minimum
fee. In normal use, Amazon Glacier has three pricing components: storage (per
GB per month), data transfer out (per GB per month), and requests (per
thousand UPLOAD and RETRIEVAL requests per month).
Page 8 of 50
Note that Amazon Glacier is designed with the expectation that retrievals are
infrequent and unusual, and data will be stored for extended periods of time.
You can retrieve up to 5% of your average monthly storage (pro-rated daily) for
free each month. If you choose to retrieve more than this amount of data in a
month, you are charged an additional (per GB) retrieval fee. There is also a pro-
rated charge (per GB) for items deleted prior to 90 days.
Pricing information can be found at Amazo n Glacier P ricing.6

Amazon Glacier scales to meet your growing and often unpredictable storage
requirements. A single archive is limited to 4 TBs, but there is no limit to the
total amount of data you can store in the service. Whether you’re storing
petabytes or gigabytes, Amazon Glacier automatically scales your storage up or
down as needed.
Interfaces
There are two ways to use Amazon Glacier, each with its own set of interfaces.
The Amazon Glacier APIs provide both management and data operations.
First, Amazon Glacier provides a native, standards-based REST web services

interface, as well as Java and .NET SDKs. The AWS Management Console or
the Amazon Glacier APIs can be used to create vaults to organize the archives in
Amazon Glacier. You can then use the Amazon Glacier APIs to upload and
retrieve archives, monitor the status of your jobs and also configure your vault
to send you a notification via Amazon Simple Notification Service (SNS) when
your jobs complete.7
Second, Amazon Glacier can be used as a storage class in Amazon S3 by using

object lifecycle management to provide automatic, policy-driven archiving from
Amazon S3 to Amazon Glacier. You simply set one or more lifecycle rules for an
Amazon S3 bucket, defining what objects should be transitioned to Amazon
Glacier and when. You can specify an absolute or relative time period (including
0 days) after which the specified Amazon S3 objects should be transitioned to
Amazon Glacier. A new RESTORE operation has been added to the Amazon S3
API, and the retrieval process takes the same 3-5 hours. On retrieval, a copy of
the retrieved object is placed in Amazon S3 RRS storage for a specified retention
period; the original archived object remains stored in Amazon Glacier. For more
Page 9 of 50
information on how to use Amazon Glacier from Amazon S3, see the Object
Lifecycle Management section of the Amazon S3 Developer Guide.8
Note that when using Amazon Glacier as a storage class in Amazon S3, you use
the Amazon S3 APIs, and when using “native” Amazon Glacier, you use the
Amazon Glacier APIs. Objects archived to Amazon Glacier via Amazon S3 can
only be listed and retrieved via the Amazon S3 APIs or the AWS Management
Console—they are not visible as archives in an Amazon Glacier vault.
Anti-Patterns
Amazon Glacier has the following anti-patterns:
• Rapidly changing data—Data that must be updated very frequently

might be better served by a storage solution with lower read/write
latencies, such as Amazon EBS or a database.
• Real time access—Data stored in Amazon Glacier is not available in real

time. Retrieval jobs typically require 3-5 hours to complete, so if you
need immediate access to your data, Amazon S3 is a better choice.
Amazon Elastic Block Store (Amazon EBS)

Volumes
Amazon Elastic Block Store (Amazon EBS) volumes provide durable block-level
storage for use with Amazon EC2 instances (virtual machines).9 Amazon EBS
volumes are off-instance, network-attached storage (NAS) that persists
independently from the running life of a single Amazon EC2 instance. After an
Amazon EBS volume is attached to an instance, you can use it like a physical
hard drive, typically by formatting it with the file system of your choice and
using the file I/O interface provided by the instance operating system. You can
use an Amazon EBS volume to boot an Amazon EC2 instance (Amazon EBS-root
AMIs only), and you can attach multiple Amazon EBS volumes to a single
Amazon EC2 instance. Note, however, that any single Amazon EBS volume may
be attached to only one Amazon EC2 instance at any point in time.
Amazon EBS also provides the ability to create point-in-time snapshots of

volumes, which are persisted to Amazon S3. These snapshots can be used as the
starting point for new Amazon EBS volumes, and to protect data for long-term
Page 10 of 50
durability. The same snapshot can be used to instantiate as many volumes as you
wish. These snapshots can be copied across AWS regions, making it easier to
leverage multiple AWS regions for geographical expansion, data center migration
and disaster recovery. Sizes for Amazon EBS volumes range from 1 GB to 1 TB,
and are allocated in 1 GB increments.

Amazon EBS is meant for data that changes relatively frequently and requires
long-term persistence. Amazon EBS is particularly well-suited for use as the
primary storage for a database or file system, or for any applications that require
access to raw block-level storage. Amazon EBS Provisioned IOPS volumes (see
next section) are particularly well-suited for use with databases applications that
require a high and consistent rate of random disk reads and writes.
Performance
Amazon EBS provides two volume types: standard volumes and Provisioned
IOPS volumes. They differ in performance characteristics and pricing model,
allowing you to tailor your storage performance and cost to the needs of your
applications. You can attach and stripe across multiple volumes of either type to
increase the I/O performance available to your Amazon EC2 applications.
Standard volumes offer cost effective storage for applications with moderate or
bursty I/O requirements. Standard volumes are designed to deliver
approximately 100 input/output operations per second (IOPS) on average with a
best effort ability to burst to hundreds of IOPS. Standard volumes are also well
suited for use as boot volumes, where the burst capability provides fast instance
start-up times.
Provisioned IOPS volumes are designed to deliver predictable, high performance

for I/O intensive workloads such as databases. With Provisioned IOPS, you
specify an IOPS rate when creating a volume, and then Amazon EBS provisions
that rate for the lifetime of the volume. Amazon EBS currently supports up to
2,000 IOPS per Provisioned IOPS volume. You can stripe multiple volumes
together to deliver thousands of IOPS per Amazon EC2 instance to your
application.
Because Amazon EBS volumes are network-attached devices, other network I/O
performed by the instance, as well as the total load on the shared network, can
Page 11 of 50
affect individual Amazon EBS volume performance. To enable your Amazon EC2
instances to fully utilize the Provisioned IOPS on an Amazon EBS volume, you
can launch selected Amazon EC2 instance types as Amazon EBS-optimized
instances. Amazon EBS-optimized instances deliver dedicated throughput
between Amazon EC2 and Amazon EBS, with options between 500 Mbps and
1,000 Mbps depending on the instance type used. When attached to Amazon
EBS-optimized instances, Provisioned IOPS volumes are designed to deliver
within 10% of the Provisioned IOPS performance 99.9% of the time.
The combination of Amazon EC2 and Amazon EBS enables you to use many of
the same disk performance optimization techniques that you would use with on-
premises servers and storage. For example, by attaching multiple Amazon EBS
volumes to a single Amazon EC2 instance, you can partition the total application
I/O load by allocating one volume for database log data, one or more volumes
for database file storage, and other volumes for file system data. Each separate
Amazon EBS volume can be configured as Amazon EBS standard or Amazon
EBS Provisioned IOPS as needed.
Alternatively, you could stripe your data across multiple similarly-provisioned

Amazon EBS volumes using RAID 0 or logical volume manager software, thus
aggregating available IOPs, total volume throughput, and total volume size.

Amazon EBS volumes are designed to be highly available and reliable. Amazon
EBS volume data is replicated across multiple servers in a single Availability
Zone to prevent the loss of data from the failure of any single component. The
durability of your Amazon EBS volume depends on both the size of your volume
and the amount of data that has changed since your last snapshot. Amazon EBS
snapshots are incremental, point-in-time backups, containing only the data
blocks changed since the last snapshot. Amazon EBS volumes that operate with
20 GB or less of modified data since their most recent snapshot can expect an
annual failure rate (AFR) between 0.1% and 0.5%. Amazon EBS volumes with
more than 20 GB of unmodified data since the last snapshot should expect
higher failure rates that are roughly proportional to the increase in modified
data.
To maximize both durability and availability of their Amazon EBS data, you
should create snapshots of your Amazon EBS volumes frequently. (For data
Page 12 of 50
consistency, it is a best practice to briefly pause any writes to the volume, or

unmount the volume, while the snapshot command is issued. You can then
safely continue to use the volume while the snapshot is pending completion.) In
the event that your Amazon EBS volume does fail, all snapshots of that volume
will remain intact, and will allow you to recreate your volume from the last
snapshot point. Because an Amazon EBS volume is created in a particular
Availability Zone, the volume will be unavailable if the Availability Zone itself is
unavailable. An Amazon EBS snapshot of a volume, however, is available
across all the Availability Zones within a region, and you can use an Amazon
EBS snapshot to create one or more new Amazon EBS volumes in any
Availability Zone in the region. Amazon EBS snapshots can also be copied from
one region to another, and can easily be shared with other user accounts. Thus,
Amazon EBS snapshots provides an easy-to-use disk clone or disk image
mechanism for backup, sharing, and disaster recovery.
Cost Model
With Amazon EBS, you pay only for what you use. Amazon EBS pricing has three
components: provisioned storage, I/O requests, and snapshot storage. Amazon
EBS standard volumes are charged per GB-month of provisioned storage and
per million I/O requests. Amazon EBS Provisioned IOPS volumes are charged
per GB-month of provisioned storage and per Provisioned IOPS-month. For
both volume types, Amazon EBS snapshots are charged per GB-month of data
stored. Amazon EBS snapshot copy is charged for the data transferred between
regions, and for the standard Amazon EBS snapshot charges in the destination
region. It’s important to remember that for Amazon EBS volumes, you are
charged for provisioned (allocated) storage, whether or not you actually use it.
For Amazon EBS snapshots, you are charged only for storage actually used
(consumed). Note that Amazon EBS snapshots are incremental and compressed,
so the storage used in any snapshot is generally much less than the storage
consumed on an Amazon EBS volume.
Note that there is no charge for transferring information among the various AWS
storage offerings (i.e., Amazon EC2 instance with Amazon EBS, Amazon S3,
Amazon RDS, and so on) as long as they are within the same AWS region.
Pricing information for Amazon EBS can be found at Am azo n EC2 Pricing.10
Page 13 of 50

Using the AWS Management Console or the APIs, Amazon EBS volumes can
easily and rapidly be provisioned and released to scale in and out with your total
storage demands. While individual Amazon EBS volumes cannot be resized, if
you find that you need additional storage, you have two ways to expand the
amount of Amazon EBS space available for your Amazon EC2 instance. The
simplest approach is to create and attach a new Amazon EBS volume and begin
using it together with your existing ones. However, if you need to expand the size
of a single Amazon EBS volume, you can effectively resize a volume using a
snapshot:
1. Detach the original Amazon EBS volume.

2. Create a snapshot of the original Amazon EBS volume’s data into Amazon
S3.
3. Create a new Amazon EBS volume from the snapshot, but specify a larger
size than the original volume.
4. Attach the new, larger volume to your Amazon EC2 instance in place of
the original. (In many cases, an OS-level utility must also be used to
expand the file system.)
5. Delete the original Amazon EBS volume.
Interfaces
Amazon offers management APIs for Amazon EBS in both SOAP and REST
formats. These are used to create, delete, describe, attach, and detach Amazon
EBS volumes for your Amazon EC2 instances; to create, delete, and describe
snapshots from Amazon EBS to Amazon S3; and to copy snapshots from one
region to another. If you prefer to work with a graphical tool, the AWS
Management Console gives you all the capabilities of the API in an easy-to-use
browser interface. Regardless of how you create your Amazon EBS volume, note
that all storage is allocated at the time of volume creation, and that you are
charged for this allocated storage even if you don’t write data to it.
There is no AWS data API for Amazon EBS. Instead, Amazon EBS presents a
block-device interface to the Amazon EC2 instance. That is, to the Amazon EC2
instance, an Amazon EBS volume appears just like a local disk drive. To write to
Page 14 of 50
and read data from Amazon EBS volumes, you therefore use the native file
system I/O interfaces of your chosen operating system.
Anti-Patterns
As described previously, Amazon EBS is ideal for information that needs to be
persisted beyond the life of a single Amazon EC2 instance. However, in certain
situations other AWS storage options may be more appropriate. Amazon EBS
has the following anti-patterns:
• Temporary storage—If you are using Amazon EBS for temporary storage
(such as scratch disks, buffers, queues, and caches), consider using local
instance store volumes, Amazon SQS, or ElastiCache (Memcached or
Redis).
• Highly-durable storage—If you need very highly-durable storage, use

Amazon S3 or Amazon Glacier. Amazon S3 standard storage is designed
for 99.999999999% annual durability per object. In contrast, Amazon
EBS volumes with less than 20 GB of modified data since the last
snapshot are designed for between 99.5% and 99.9% annual durability;
volumes with more modified data can be expected to have proportionally
lower durability.
• Static data or web content—If your data doesn’t change that often,
Amazon S3 may represent a more cost-effective and scalable solution for
storing this fixed information. Also, web content served out of Amazon
EBS requires a web server running on Amazon EC2, while you can
deliver web content directly out of Amazon S3.
Amazon EC2 Instance Store Volumes

Amazon EC2 instance store volumes (also called ephemeral drives) provide
temporary block-level storage for many Amazon EC2 instance types.11 This
storage consists of a preconfigured and pre-attached block of disk storage on the
same physical server that hosts the Amazon EC2 instance.12 . The amount of this
disk storage varies by Amazon EC2 instance type. In those Amazon EC2 instance
families that provide instance storage, larger instances tend to provide both more
and larger instance store volumes. Note that some instance types, such as the
micro instances (t1) use Amazon EBS storage only, with no instance storage
provided.
Page 15 of 50
In addition, the storage-optimized Amazon EC2 instance family provides special-

purpose instance storage targeted to specific use cases. HI1 instances provide
very fast solid-state drive (SSD)-backed instance storage capable of supporting
over 120,000 random read IOPS, and are optimized for very high random I/O
performance and low cost per IOPS. In contrast, HS1 instances are optimized for
very high storage density, low storage cost, and high sequential I/O
performance.

In general, local instance store volumes are ideal for temporary storage of
information that is continually changing, such as buffers, caches, scratch data,
and other temporary content, or for data that is replicated across a fleet of
instances, such as a load-balanced pool of web servers. Amazon EC2 instance
storage is well-suited for this purpose. It consists of the virtual machine’s boot
device (for instance store AMIs only), plus one or more additional volumes that
are dedicated to the Amazon EC2 instance (for both Amazon EBS AMIs and
instance store AMIs). This storage is usable only from a single Amazon EC2
instance during its lifetime. Note that, unlike Amazon EBS volumes, instance
store volumes cannot be detached or attached to another instance.
High I/O and high storage provide Amazon EC2 instance storage targeted to
specific use cases. High I/O instances provide instance store volumes backed by
SSD, and are ideally suited for many high performance database workloads.
Example applications include NoSQL databases like Cassandra and MongoDB.
High storage instances support much higher storage density per Amazon EC2
instance, and are ideally suited for applications that benefit from high sequential
I/O performance across very large datasets. Example applications include data
warehouses, Hadoop storage nodes, seismic analysis, cluster file systems, etc.
Note that applications using instance storage for persistent data generally
provide data durability through replication, or by periodically copying data to
durable storage.
Performance
The non-SSD-based instance store volumes in most Amazon EC2 instance
families have performance characteristics similar to standard Amazon EBS
volumes. Because the Amazon EC2 instance virtual machine and the local
instance store volumes are located in the same physical server, interaction with
Page 16 of 50
this storage is very fast, particularly for sequential access. To increase aggregate
IOPS, or to improve sequential disk throughput, multiple instance store volumes
can be grouped together using RAID 0 (disk striping) software. Because the
bandwidth to the disks is not limited by the network, aggregate sequential
throughput for multiple instance volumes can be higher than for the same
number of Amazon EBS volumes.
The SSD instance store volumes in the Amazon EC2 high I/O instances provide
from tens of thousands to hundreds of thousands of low-latency, random 4 KB
random IOPS. Because of the I/O characteristics of SSD devices, write
performance can be variable. For more information, see High I/O Instances in
the Amazon EC2 User Guide.13
The instance store volumes on Amazon EC2 high storage instances provide very
high storage density and high sequential read and write performance. High
storage instances are capable of delivering 2.6 GB/sec of sequential read and
write performance when using a block size of 2 MB. For more information, see
High Storage Instances in the Amazon EC2 User Guide.14

Amazon EC2 local instance store volumes are not intended to be used as durable
disk storage. Unlike Amazon EBS volume data, data on instance store volumes
persists only during the life of the associate Amazon EC2 instance. This means
that data on instance store volumes is persistent across orderly instance reboots,
but if the Amazon EC2 instance is stopped and re-started, terminates, or fails, all
data on the instance store volumes is lost. For more information on the lifecycle
of an Amazon EC2 instance, see Instance Lifecycle.15
You should not use local instance store volumes for any data that must persist
over time, such as permanent file or database storage, without providing for data
persistence by replicating your data, or by periodically copying data to durable
storage such as Amazon EBS or Amazon S3. Note that this also applies to the
special-purpose SSD and high-density instance store volumes in the high I/O and
high storage instance types.
Cost Model
The cost of the Amazon EC2 instance includes any local instance store volumes,
if the instance type provides them. While there is no additional charge for data
Page 17 of 50
storage on local instance store volumes, note that data transferred to and from
Amazon EC2 instance store volumes from other Availability Zones or outside of
an Amazon EC2 region may incur data transfer charges, and additional charges
will apply for use of any persistent storage, such as Amazon S3, Amazon Glacier,
Amazon EBS volumes, and Amazon EBS snapshots. Pricing information for
Amazon EC2, Amazon EBS, and data transfer can be found at Amazon EC2
Pricing.16

The number and storage capacity of Amazon EC2 local instance store volumes
are fixed and defined by the instance type.
While you can’t increase or decrease the number of instance store volumes on a
single Amazon EC2 instance, this storage is still scalable and elastic, in that you
can scale the total amount of instance store up or down by increasing or
decreasing the number of running Amazon EC2 instances.
Local instance store volumes are tied to a particular Amazon EC2 instance, and
are fixed in number and size for a given Amazon EC2 instance type, so the
scalability and elasticity of this storage is tied to the number of Amazon EC2
instances.
However, you can achieve full storage elasticity by including one of the other
suitable storage options, such as Amazon S3 or Amazon EBS, in your Amazon
EC2 storage strategy.
Interfaces
There is no separate management API for Amazon EC2 instance store volumes.
Instead, instance store volumes are specified using the block device mapping
feature of the Amazon EC2 API and the AWS Management Console. You cannot
create or destroy instance store volumes, but you can control whether or not they
are exposed to the Amazon EC2 instance, and what device name is used.
There is also no separate data API for instance store volumes. Just like Amazon
EBS volumes, instance store volumes present a block-device interface to the
Amazon EC2 instance. That is, to the Amazon EC2 instance, an instance store
volume appears just like a local disk drive. To write to and read data from
Page 18 of 50
instance store volumes, you therefore use the native file system I/O interfaces of
your chosen operating system.
Note that in some cases, a local instance store volume device will be attached to
the Amazon EC2 instance upon launch, but must be formatted with an
appropriate file system and mounted before use. Also, keep careful track of your
block device mappings. There is no simple way for an application running on an
Amazon EC2 instance to determine which block device is an instance store
(ephemeral) volume and which is an Amazon EBS (persistent) volume.
Anti-Patterns
Amazon EC2 local instance store volumes are fast, free (that is, included in the
price of the Amazon EC2 instance) “scratch volumes” best suited for storing
temporary data that can easily be regenerated, or data that is replicated for
durability. In many situations, however, other AWS storage options may be more
appropriate. Amazon EC2 instance store volumes have the following anti-
patterns:
• Persistent storage—If you need persistent virtual disk storage similar to

a physical disk drive for files or other data that must persist longer than
the lifetime of a single Amazon EC2 instance, Amazon EBS volumes or
Amazon S3 are more appropriate.
• Relational database storage—In most cases, relational databases

require storage that persists beyond the lifetime of a single Amazon EC2
instance, making Amazon EBS volumes the natural choice.
• Shared storage—Instance store volumes are dedicated to a single

Amazon EC2 instance, and cannot be shared with other systems or users.
If you need storage that can be detached from one instance and attached
to a different instance, or if you need the ability to share data easily,
Amazon S3 or Amazon EBS volumes are the better choice.
• Snapshots—If you need the convenience, long-term durability,

availability, and shareability of point-in-time disk snapshots, Amazon
EBS volumes are a better choice.
Page 19 of 50
AWS Import/Export
AWS Import/Export accelerates moving large amounts of data into and out of
AWS using portable storage devices for transport. 17 AWS transfers your data
directly onto and off of storage devices using Amazon’s high-speed internal
network and bypassing the Internet. For significant datasets, AWS
Import/Export is often faster than Internet transfer and more cost effective
than upgrading your connectivity.
AWS Import/Export supports importing and exporting data into and out of
several types of AWS storage, including Amazon EBS snapshots, Amazon S3
buckets, and Amazon Glacier vaults.

AWS Import/Export is ideal for transferring large amounts of data in and out of
the AWS cloud, especially in cases where transferring the data over the Internet
would be too slow or too costly. In general, if loading your data over the Internet
would take a week or more, you should consider using AWS Import/Export.
Common use cases include initial data upload to AWS, content distribution or
regular data interchange to/from your customers or business associates, transfer
to Amazon S3 or Amazon Glacier for off-site backup and archival storage, and
quick retrieval of large backups from Amazon S3 or Amazon Glacier for disaster
recovery.
Performance
Each AWS Import/Export station is capable of loading data at over 100 MB per
second, but in most cases the rate of the data load will be bounded by a
combination of the read or write speed of your portable storage device and, for
Amazon S3 data loads, the average object (file) size.

Once the data is imported to AWS, the durability and availability characteristics
of the target storage applies. Amazon EBS snapshots, Amazon S3, and Amazon
Glacier are all designed for 99.999999999% (11 nines) durability. For more
details, see the sections on each individual service in this whitepaper.
Page 20 of 50
Cost Model
With AWS Import/Export, you pay only for what you use. AWS Import/Export
has three pricing components: a per-device fee, a data load time charge (per
data-loading-hour), and possible return shipping charges (for expedited
shipping, or shipping to destinations not local to that AWS Import/Export
region). For the destination storage, the standard Amazon EBS snapshot,
Amazon S3, and Amazon Glacier request and storage pricing applies. Pricing
information can be found at AWS Import/Export Pricing.18

The total amount of data you can load using AWS Import/Export is limited only
by the capacity of the devices that you send to AWS. For Amazon S3, individual
files will be loaded as objects in Amazon S3, and may range up to 5 terabytes in
size. For Amazon Glacier, individual devices will be loaded as a single archive,
and may range up to 4 terabytes in size.
The aggregate total amount of data that can be imported is virtually unlimited.
Interfaces
To upload or download data, you must create and submit an AWS
Import/Export job for each storage device shipped. Each job request requires a
manifest file, a YAML-formatted text file that contains a set of key-value pairs
that supply the required information—such as your device ID, secret access key,
and return address—necessary to complete the job.
Jobs can be created using a command line tool (the AWS Import/Export Web
Service Tool), the AWS SDK for Java, or a native REST API. The job request is
tied to the storage device through a signature file in the root directory (for
Amazon S3 import jobs), or by a barcode taped to the device (for Amazon EBS
and Amazon Glacier jobs).
Anti-Patterns
AWS Import/Export is optimal for large data that would take too long to load
over the Internet, so the anti-pattern is simply data that is more easily
transferred over the Internet. If your data can be transferred over the Internet in
less than one week, AWS Import/Export may not be the ideal solution.
Page 21 of 50
AWS Storage Gateway

AWS Storage Gateway is a service that connects an on-premises software
appliance with cloud-based storage to provide seamless and secure integration
between an organization’s on-premises IT environment and AWS’s storage
infrastructure.19 The service enables you to securely store data to the AWS cloud
for scalable and cost-effective storage. AWS Storage Gateway supports industry-
standard storage protocols that work with your existing applications. It provides
low-latency performance by maintaining frequently accessed data on-premises
while securely storing all of your data encrypted in Amazon S3. For disaster
recovery scenarios, it can serve as a cloud-hosted solution, together with
Amazon EC2, that mirrors your entire production environment.
AWS Storage Gateway’s software appliance is available for download as a virtual

machine (VM) image that you install on a host in your datacenter. Once you’ve
installed your gateway and associated it with your AWS account through our
activation process, you can use the AWS Management Console to create either
gateway-cached or gateway-stored volumes that can be mounted as iSCSI
devices by your on-premises applications.
Gateway-cached volumes allow you to utilize Amazon S3 for your primary data,
while retaining some portion of it locally in a cache for frequently accessed data.
These volumes minimize the need to scale your on-premises storage
infrastructure, while still providing your applications with low-latency access to
their frequently accessed data. You can create storage volumes up to 32 TBs in
size and mount them as iSCSI devices from your on-premises application
servers. Data written to these volumes is stored in Amazon S3, with only a cache
of recently written and recently read data stored locally on your on-premises
storage hardware.
Gateway-stored volumes store your primary data locally, while asynchronously

backing up that data to AWS. These volumes provide your on-premises
applications with low-latency access to their entire datasets, while providing
durable, off-site backups. You can create storage volumes up to 1 TB in size and
mount them as iSCSI devices from your on-premises application servers. Data
written to your gateway-stored volumes is stored on your on-premises storage
hardware, and asynchronously backed up to Amazon S3 in the form of Amazon
EBS snapshots.
Page 22 of 50

Organizations are using AWS Storage Gateway to support a number of use cases.
These include corporate file sharing, enabling existing on-premises backup
applications to store primary backups on Amazon S3, disaster recovery, and data
mirroring to cloud-based compute resources.
Performance
As the AWS Storage Gateway VM sits between your application, Amazon S3, and
underlying on-premises storage, the performance you experience will be
dependent upon a number of factors, including the speed and configuration of
your underlying local disks, the network bandwidth between your iSCSI
initiator and gateway VM, the amount of local storage allocated to the gateway
VM, and the bandwidth between the gateway VM and Amazon S3. For gateway-
cached volumes, to provide low-latency read access to your on-premises
applications, it’s important that you provide enough local cache storage to store
your recently accessed data. Our AWS Storage Gateway documentation provides
guidance on how to optimize your environment setup for best performance,
including how to properly size your local storage.20
AWS Storage Gateway efficiently uses your Internet bandwidth to speed up the
upload of your on-premises application data to AWS. AWS Storage Gateway
only uploads data that has changed, which minimizes the amount of data sent
over the Internet. You can also use AWS Direct Connect to further increase
throughput and reduce your network costs by establishing a dedicated network
connection between your on-premises gateway and AWS.21

AWS Storage Gateway durably stores your on-premises application data by
uploading it to Amazon S3. Amazon S3 stores data in multiple facilities and on
multiple devices within each facility. Amazon S3 also performs regular,
systematic data integrity checks and is built to be automatically self-healing.
Cost Model
With AWS Storage Gateway, you pay only for what you use. AWS Storage
Gateway has four pricing components: gateway usage (per gateway per month),
snapshot storage usage (per GB per month), volume storage usage (per GB per
Page 23 of 50
month), and data transfer out (per GB per month). Pricing information can be
found at AWS Storage Gateway Pricing.22

In both gateway-cached and gateway-stored volume configurations, AWS
Storage Gateway stores data in Amazon S3, which has been designed to offer a
very high level of scalability and elasticity automatically. Unlike a typical file
system that encounters issues when storing large number of files in a directory,
Amazon S3 supports a virtually unlimited number of files in any bucket. Also,
unlike a disk drive that has a limit on the total amount of data that can be stored
before you must partition the data across drives and/or servers, an Amazon S3
bucket can store a virtually unlimited number of bytes. You are able to store any
number of objects, and Amazon S3 will manage scaling and distributing
redundant copies of your information onto other servers in other locations in the
same region, all using Amazon’s high-performance infrastructure.
Interfaces
The AWS Management Console can be used to download the AWS Storage
Gateway VM image. You can then select between a gateway-cached or gateway-
stored configuration, activate your on-premises by associating your gateway’s IP
Address with your AWS account, select an AWS region, and create AWS Storage
Gateway volumes and attach these volumes as iSCSI devices to your on-
premises application servers.
You can begin using the AWS Storage Gateway in just a few steps. To get started,
you simply:
1. Download the AWS Storage Gateway VM image from the AWS

Management Console.23
2. Select between a gateway-cached or gateway-stored configuration. With
gateway-cached volumes, you store your primary data in Amazon S3, and
retain frequently accessed data locally. With gateway-stored volumes, you
store your primary data locally, and upload backups to Amazon S3.
3. Allocate local storage to your installed on-premises gateway from direct-
attached storage (DAS), network attached storage (NAS), or storage area
network (SAN) storage. For gateway-cached configurations, this local
Page 24 of 50
storage will be used for your frequently accessed data. For gateway-stored
configurations, it will be used for your primary data.
4. Activate your on-premises by associating your gateway’s IP Address with
your AWS account and select an AWS region for your gateway to store
uploaded data.
5. Use the AWS Management Console to create AWS Storage Gateway
volumes and attach these volumes as iSCSI devices to your on-premises
application servers.
By following these steps, you can begin using your existing on-premises
applications to seamlessly store data in Amazon S3. These applications can now
write data to their attached AWS Storage Gateway volumes. Your application data
will either be stored directly in Amazon S3 for substantial cost savings on
primary storage (gateway-cached volumes), or will be stored locally and backed
up to Amazon S3 for durable and cost-effective backups (gateway-stored
volumes).
Anti-Patterns
AWS Storage Gateway has the following anti-patterns:
• Database storage—Amazon EC2 instances using Amazon EBS volumes

are a natural choice for database storage and workloads.
Amazon CloudFront
Amazon CloudFront content delivery network (CDN) is a web service for content
delivery.24 Amazon CloudFront makes your website's dynamic, static, and
streaming content available from a global network of edge locations. When a
visitor requests a file from your website, he or she is invisibly redirected to a
copy of the file at the nearest edge location, which results in faster download
times than if the visitor had accessed the content from a data center farther
away. Amazon CloudFront caches content at edge locations for a period of time
that you specify.
Amazon CloudFront supports all files that can be served over HTTP. This
includes dynamic web pages, such as HTML or PHP pages, any popular static
files that are a part of your web application, such as website images, audio,
video, media files or software downloads. For on-demand media files, you can
Page 25 of 50
also choose to stream your content using Real-Time Messaging Protocol

(RTMP) delivery. Amazon CloudFront also supports delivery of live media over
HTTP.
Amazon CloudFront is optimized to work with other Amazon Web Services, like
Amazon S3, Amazon EC2, Amazon Elastic Load Balancing, and Amazon Route
53. Amazon CloudFront also works seamlessly with any non-AWS origin server,
which stores the original, definitive versions of your files.

Amazon CloudFront is ideal for distribution of frequently-accessed static
content that benefits from edge delivery—like popular website images, videos,
media files or software downloads. Amazon CloudFront can also be used to
deliver dynamic web applications over HTTP. These applications may include
static content, dynamic content, or a whole site with a mixture of the two.
Amazon CloudFront is also commonly used to stream audio and video files to
web browsers and mobile devices.
Performance
Amazon CloudFront is designed for low-latency and high-bandwidth delivery of
content. Amazon CloudFront speeds up the distribution of your content by
routing end users to the edge location that can best serve the end user's request
in a worldwide network of edge locations. Typically, requests are routed to the
nearest Amazon CloudFront edge location in terms of latency. This dramatically
reduces the number of networks that your users' requests must pass through
and improves performance. End users get both lower latency—the time it takes
to load the first byte of the object—and higher sustained data transfer rates
needed to deliver popular objects to end user at scale.

Because a CDN is an edge cache, Amazon CloudFront does not provide durable
storage. Durable file storage is provided by the origin server, such as Amazon S3
or a web server running on Amazon EC2. Amazon CloudFront provides high
availability by using a distributed global network of edge locations. Origin
requests from the edge locations to AWS origin servers (e.g., Amazon EC2,
Amazon S3, etc.) are carried over network paths that Amazon constantly
monitors and optimizes for both availability and performance. This edge
network provides increased reliability and availability because there is no longer
Page 26 of 50
a central point of failure. Copies of your files are now held in edge locations
around the world.
Cost Model
With Amazon CloudFront, there are no long-term contracts or required
minimum monthly commitments—you pay only for as much content as you
actually deliver through the service. Amazon CloudFront has two pricing
components: regional data transfer out (per GB) and requests (per 10,000).
Note that it is often cheaper (as well as faster) to deliver popular content from
Amazon S3 through Amazon CloudFront rather than directly from Amazon S3.
Pricing information can be found at Amazon CloudFront Pricing.25

Amazon CloudFront is designed to provide seamless scalability and elasticity.
You can easily start very small, and grow to massive numbers of global
connections. With Amazon CloudFront, you don’t need to worry about
maintaining expensive web server capacity to meet the demand from potential
traffic spikes for your content. The service automatically responds as demand
increases or decreases without any intervention from you. Amazon CloudFront
also uses multiple layers of caching at each edge location and collapses
simultaneous requests for the same object before contacting your origin server.
These optimizations further help reduce the need to scale your origin
infrastructure as your website becomes more popular.
Interfaces
There are several ways to manage and configure Amazon CloudFront. The AWS
Management Console provides an easy way to manage Amazon CloudFront. All
the features of the Amazon CloudFront API are supported. For example, you can
enable or disable distributions, configure CNAMEs, and enable end-user
logging. You can also use the Amazon CloudFront command line tools, the
native REST API, or one of the supported SDKs.
There is no data API for Amazon CloudFront, no command to preload data.

Instead, data is automatically pulled into Amazon CloudFront edge locations on
the first access of an object from that location.
Page 27 of 50
Anti-Patterns
Amazon CloudFront is optimal for delivery of popular static or dynamic.
However, in a number of situations Amazon CloudFront is not the optimal
solution. Amazon CloudFront has the following anti-patterns:
• Programmatic cache invalidation—While Amazon CloudFront

supports cache invalidation, AWS recommends using object versioning
rather than programmatic cache invalidation.
• Infrequently requested data—It may be better to serve infrequently-

accessed data directly from the origin server, avoiding the additional cost
of origin fetches for data that is not likely to be reused at the edge.
Amazon Simple Queue Service (Amazon SQS)

Amazon Simple Queue Service (Amazon SQS) provides a reliable, highly
scalable, hosted message queuing service for temporary storage and delivery of
short (up to 64 KB) text-based data messages.26 An Amazon SQS queue is a
temporary data repository for messages that are waiting for processing
(typically a message produced by one application component and waiting to be
consumed by another). Amazon SQS messages can be sent and received by
servers or distributed application components within the Amazon EC2
environment or anywhere on the Internet. Amazon SQS supports a virtually
unlimited number of queues and supports unordered, at-least-once delivery of
messages.
While Amazon SQS and other message queuing services are usually thought of
as an asynchronous communication protocols, Amazon SQS can also be viewed
as a class of temporary but durable data storage for many classes of
applications. Use of Amazon SQS as temporary storage can minimize the use of
other storage mechanisms, such as temporary disk files.

Amazon SQS is ideally suited to any scenario where multiple application
components must communicate and coordinate their work in a loosely coupled
manner. This occurs particularly in producer-consumer scenarios where some
components may work faster or slower than others, or where the number of
interacting components changes with time or load. Amazon SQS can serve as
Page 28 of 50
the “software glue” that enables components to communicate reliably without

being tightly coupled or highly dependent upon synchronous operation, or on a
fixed number of components.
A classic use of Amazon SQS is to coordinate a multi-step processing pipeline,

where each message is associated with a task that must be processed. Each task
is described by an Amazon SQS message indicating the task to be done and a
pointer to the task data in Amazon S3.
To illustrate, suppose you have a number of image files to encode. In an Amazon

SQS worker queue, you create an Amazon SQS message for each file specifying
the command (jpeg-encode) and the location of the file in Amazon S3. A pool of
Amazon EC2 instances running the needed image processing software does the
following:
1. Asynchronously pulls the task messages from the queue

2. Retrieves the named file
3. Processes the conversion
4. Writes the image back to Amazon S3
5. Writes a “task complete” message to another queue
6. Deletes the original task message
7. Checks for more messages in the worker queue
Use of the Amazon SQS queue enables the number of worker instances to scale
up or down, and also enable the processing power of each single worker
instance to scale up or down, to suit the total workload, without any application
changes.
Performance
Amazon SQS is a distributed queuing system that is optimized for horizontal
scalability, not for single-threaded sending or receiving speeds. A single client
can send or receive Amazon SQS messages at a rate of about 5 to 50 messages
per second. Higher receive performance can be achieved by requesting multiple
messages (up to 10) in a single call. It may take several seconds before a
message that has been to a queue is available to be received.
Page 29 of 50

By design, messages stored in Amazon SQS are highly durable but temporary.
To prevent messages from being lost or becoming unavailable, all messages are
stored redundantly across multiple servers and data centers. Message retention
time is configurable on a per-queue basis, from a minimum of one hour to a
maximum of 14 days. Messages are retained in a queue until they are explicitly
deleted, or until they are automatically deleted upon expiration of the retention
time.
Cost Model
With Amazon SQS, you pay only for what you use and there is no minimum fee.
To get started and to support simple applications, Amazon SQS provides a free
tier of service which provides 100,000 requests per month at no charge. Beyond
the free tier, Amazon SQS pricing is based on number of requests (priced per
10,000 requests) and the amount of data transferred in and out (priced per GB
per month). Pricing information can be found at Amazon SQS Pricing.27

Amazon SQS is both highly elastic and massively scalable. It is designed to
enable a virtually unlimited number of computers to read and write a virtually
unlimited number of messages at any time. It supports virtually unlimited
numbers of queues and messages per queue for any user.
Interfaces
Amazon SQS can be accessed through an HTTP Query web services interface, as
well as through SDKs for Java, PHP, Ruby, and .NET. The Amazon SQS APIs
provides both management and data interfaces. Five APIs make it easy for
developers to get started with Amazon SQS: CreateQueue, SendMessage,
ReceiveMessage, ChangeMessageVisibility, and DeleteMessage. Additional APIs
provide advanced functionality.
Anti-Patterns
Amazon SQS has the following anti-patterns:
• Binary or large messages—Amazon SQS messages must be text, and

can be a maximum of 64 KB in length. If the data you need to store in a
queue exceeds this length, or is binary, it is best to use Amazon S3 or
Page 30 of 50
Amazon RDS to store the large or binary data, and store a pointer to the
data in Amazon SQS.
• Long-term storage—If message data must be stored for longer than 14

days, Amazon S3 or some other storage mechanism is more appropriate.
• High-speed message queuing or very short tasks—If your application

requires a very high-speed message send and receive response from a
single producer or consumer, use of Amazon DynamoDB or a message-
queuing system hosted on Amazon EC2 may be more appropriate.
Amazon Relational Database Service (Amazon

RDS)
Amazon Relational Database Service (Amazon RDS) is a web service that
provides the capabilities of MySQL, Oracle, or Microsoft SQL Server relational
database as a managed, cloud-based service.28 It also eliminates much of the
administrative overhead associated with launching, managing, and scaling your
own relational database on Amazon EC2 or in another computing environment.

Amazon RDS is ideal for existing applications that rely on MySQL, Oracle, or
SQL Server traditional relational database engines. Since Amazon RDS offers
full compatibility and direct access to native database engines, most code,
libraries, and tools designed for these databases should work unmodified with
Amazon RDS. Amazon RDS is also optimal for new applications with structured
data that requires more sophisticated querying and joining capabilities than
that provided by Amazon’s NoSQL database offering, Amazon DynamoDB.
Performance
Amazon RDS delivers high performance through a combination of configurable
instances running on Amazon’s proven, world-class infrastructure with fully-
automated maintenance and backup operations. Available database
configurations range from a small instance (64-bit platform with 1.7 GB of RAM
and 1 Amazon EC2 compute unit (ECU)) up to a quadruple extra-large instance
(64-bit platform with 68 GB of RAM and 26 ECUs).
Page 31 of 50
Amazon RDS Provisioned IOPS provides a high-performance storage option

designed to deliver fast, predictable, and consistent performance for I/O
intensive transactional database workloads. When creating a new DB instance
using the Amazon RDS Provisioned IOPS storage, you can specify the IOPS your
instance needs from 1,000 IOPS to 30,000 IOPS and Amazon RDS provisions
that IOPS rate for the lifetime of the instance.

As with other relational databases on Amazon EC2, Amazon RDS leverages
Amazon EBS volumes as its data store. For enhanced durability, Amazon RDS
offers two types of database backups that are replicated across multiple
Availability Zones: automated DB instance backups and user-initiated database
snapshots. If you enable automated DB instance backups, Amazon RDS will
automatically perform a full daily backup of your data during the specified
backup window, and will also capture DB transaction logs. These automated
backups are provided at no additional charge, can be retained for up to eight
days, and can be used to do a point-in-time restore to any point from the start of
the retention period to about the last five minutes from current time. DB
snapshots are user-initiated, can be created at any time, and are kept until
explicitly deleted. DB snapshots allow you to restore your database to a known
state.
The Amazon RDS Multi-AZ deployment feature enhances both the durability
and the availability of your database by synchronously replicating your data
between a primary Amazon RDS DB instance and a standby instance in another
Availability Zone. In the unlikely event of a DB component failure or an
Availability Zone failure, Amazon RDS will automatically failover to the standby
(which typically takes about three minutes) and the database transactions can
be resumed as soon as the standby is promoted. The synchronous replication
helps to prevent loss of data.
Note that the synchronous replication provided by Amazon RDS Multi-AZ

deployment feature is complementary to the built-in asynchronous replication
provided by Amazon RDS read replicas. You can use either feature alone, or
both in combination.
Page 32 of 50
Cost Model
With Amazon RDS, you pay only for what you use and there is no minimum fee.
Amazon RDS offers a tiered pricing structure, based on the size of the database
instance, the deployment type (Single-AZ/Multi-AZ), and the AWS region.
Pricing for Amazon RDS is based on several factors: the DB instance hours (per
hour), the amount of provisioned database storage (per GB-month and per
million I/O requests), additional backup storage (per GB-month), and data
transfer in / out (per GB per month). Pricing information can be found at
Amazon Relational Database Service Pricing.29

Amazon RDS resources can be scaled elastically in several dimensions: database
storage size, database storage IOPS rate, database instance compute capacity,
and the number of read replicas.
Amazon RDS supports “pushbutton scaling” of both database storage and

compute resources. To scale the Amazon RDS database storage elastically, you
can use the command line tools, API, or AWS Management Console to provision
additional storage. Depending on your needs, this additional storage can be
added either immediately or during the next maintenance window. For
databases needing consistent I/O performance, you can scale the throughput of
your database by specifying the IOPS rate, from 1,000 to 30,000 IOPS, with
corresponding storage size of 100 GB to 3 TB.
Scaling computational resources up or down by switching from one DB instance

type to another is easily performed with a single API or AWS Management
Console command. For example, you might need additional compute power to
create invoices at the end of each month. It’s simple to temporarily scale up to a
quadruple extra-large instance, run the computationally intensive workloads,
and then return to a smaller, more cost-effective configuration for the
remainder of the month.
Amazon RDS for MySQL also enables you to scale out beyond the capacity of a
single database deployment for read-heavy database workloads by creating one
or more read replicas. Read replicas use MySQL’s built-in asynchronous
replication capability, and can be used in conjunction with the synchronous
replication provided by Amazon RDS Multi-AZ deployments.
Page 33 of 50
Finally, in the more general case, administrators may configure multiple

Amazon RDS instances and leverage database partitioning or sharding to spread
the workload over multiple DB instances, achieving even greater database
scalability and elasticity.
Interfaces
Amazon RDS APIs and the AWS Management Console provide a management
interface that allows you to create, delete, modify, and terminate Amazon RDS
DB instances; to create DB snapshots; and to perform point-in-time restores. To
start using Amazon RDS, you simply use the AWS Management Console or
Amazon RDS APIs to launch a database instance (DB instance), selecting the
DB engine (MySQL, Oracle or SQL Server), license type, DB instance class, and
storage capacity that best meets your needs.
There is no AWS data API for Amazon RDS. Once the database is created, AWS
provides a DNS endpoint through which you can connect to your DB instance
using your favorite database tool or programming language. Since you have
direct access to a native MySQL, Oracle, or SQL Server database engine, most
tools designed for these engines should work unmodified with Amazon RDS.
After your schema and data are in place, you interact with your information via
standard SQL, as well as JDBC and other popular APIs, and any graphical tools
that can work with relational data. There are no code changes to be made to let
your application interact with Amazon RDS. You simply replace your database
server’s address (e.g. dbserver.example.com) in you database connection string
with the public DNS endpoint (e.g. myinstance.c0cafggtpzd2.us-east-
1.rds.amazonaws.com) provided by AWS when you create the instance. This
DNS endpoint will remain the same for the lifetime of your instance, even after
failover of an Amazon RDS Multi-AZ deployment. Aside from configuring the
endpoint, everything else about your database-centric application is unchanged.
Anti-Patterns
Amazon RDS is a great solution for cloud-based fully-managed relational
database, but in a number of scenarios it may not be the right choice. Amazon
RDS has the following anti-patterns:
• Index and query-focused data—Many cloud-based solutions don’t

require advanced features found in a relational database, such as joins
and complex transactions. If your application is more oriented toward
Page 34 of 50
indexing and querying data, you may find Amazon DynamoDB to be

more appropriate for your needs.
• Numerous BLOBs—While all of the database engines provided by

Amazon RDS support binary large objects (BLOBs), if your application
makes heavy use of them (audio files, videos, images, and so on), you
may find Amazon S3 to be a better choice.
• Automated scalability—As stated previously, Amazon RDS provides

pushbutton scaling. If you need fully-automated scaling, Amazon
DynamoDB may be a better choice.
• Other database platforms—At this time, Amazon RDS provides a

MySQL, Oracle, and SQL Server databases. If you need another database
platform (such as IBM DB2, Informix, PostgreSQL, or Sybase) you need
to deploy a self-managed database on an Amazon EC2 instance by using
a relational database AMI, or by installing database software on an
Amazon EC2 instance.
• Complete control—If your application requires complete, OS-level

control of the database server, with full root or admin login privileges
(for example, to install additional third-party software on the same
server), a self-managed database on Amazon EC2 may be a better match.
Amazon DynamoDB
Amazon DynamoDB is a fast, fully-managed NoSQL database service that
makes it simple and cost-effective to store and retrieve any amount of data, and
serve any level of request traffic.30 Amazon DynamoDB helps offload the
administrative burden of operating and scaling a highly-available distributed
database cluster. This storage alternative meets the latency and throughput
requirements of highly demanding applications by providing extremely fast and
predictable performance with seamless throughput and storage scalability.
Amazon DynamoDB stores structured data in tables, indexed by primary key,

and allows low-latency read and write access to items ranging from 1 byte up to
64 KB. Amazon DynamoDB supports three data types: number, string, and
binary, in both scalar and multi-valued sets. Tables do not have fixed a schema,
so each data item can have a different number of attributes. The primary key
can either be a single-attribute hash key or a composite hash-range key. Local
Page 35 of 50
secondary indexes provide additional flexibility for querying against attributes

other than the primary key. Amazon DynamoDB provides both eventually-
consistent reads (by default), and strongly-consistent reads (optional), as well as
implicit item-level transactions for item put, update, delete, conditional
operations, and increment/decrement.
Amazon DynamoDB is integrated with other services, such as Amazon Elastic

MapReduce (Amazon EMR), Amazon Redshift, Amazon Data Pipeline, and
Amazon S3, for analytics, data warehouse, data import/export, backup, and
archive.

Amazon DynamoDB is ideal for existing or new applications that need a flexible
NoSQL database with low read and write latencies, and the ability to scale
storage and throughput up or down as needed without code changes or
downtime.
Common use cases include: mobile apps, gaming, digital ad serving, live voting
and audience interaction for live events, sensor networks, log ingestion, access
control for web-based content, metadata storage for Amazon S3 objects, e-
commerce shopping carts, and web session management. Many of these use
cases require a highly available and scalable database because downtime or
performance degradation has an immediate negative impact on an
organization’s business.
Performance
SSDs and limiting indexing on attributes provides high throughput and low
latency (single-digit milliseconds typical for average server-side response
times), and drastically reduces the cost of read and write operations. As the
datasets grow, predictable performance is required so that low-latency for the
workloads can be maintained. This predictable performance can be achieved by
defining the provisioned throughput capacity required for a given table. Behind
the scenes, the service handles the provisioning of resources to achieve the
requested throughput rate, which takes the burden away from the customer to
have to think about instances, hardware, memory, and other factors that can
affect an application’s throughput rate. Provisioned throughput capacity
reservations are elastic and can be increased or decreased on demand.
Page 36 of 50

Amazon DynamoDB has built-in fault tolerance that automatically and
synchronously replicates data across three Availability Zones in a region for
high availability and to help protect data against individual machine, or even
facility failures.
Cost Model
With Amazon DynamoDB, you pay only for what you use and there is no
minimum fee. Amazon DynamoDB has three pricing components: provisioned
throughput capacity (per hour), indexed data storage (per GB per month), data
transfer in or out (per GB per month). New customers can start using Amazon
DynamoDB for free as part of the AWS Free Usage Tier. Pricing information can
be found at Amazon DynamoDB Pricing.31

Amazon DynamoDB is both highly-scalable and elastic. There is no limit to the
amount of data you can store in an Amazon DynamoDB table, and the service
automatically allocates more storage as you store more data using the Amazon
DynamoDB write APIs. Data is automatically partitioned and re-partitioned as
needed, while the use of SSDs provides predictable low-latency response times
at any scale. The service is also elastic, in that you can simply “dial-up” or “dial-
down” the read and write capacity of a table as your needs change.
Interfaces
Amazon DynamoDB provides a low-level REST API, as well as higher-level
SDKs for Java, .NET, and PHP that wrap the low- level REST API and provide
some object-relational mapping (ORM) functions. These APIs provide both a
management and data interface for Amazon DynamoDB. The API currently
offers thirteen operations that enable table management (creating, listing,
deleting, and obtaining metadata) and working with attributes (getting, writing,
and deleting attributes; query using an index, and full scan). While standard
SQL isn’t available for Amazon DynamoDB, you may use the Amazon
DynamoDB select operation to create SQL-like queries that retrieve a set of
attributes based on criteria that you provide. You can also work with Amazon
DynamoDB using the AWS Management Console.
Page 37 of 50
Anti-Patterns
Amazon DynamoDB has the following anti-patterns:
• Prewritten application tied to a traditional relational database—If

you are attempting to port an existing application to the AWS cloud, and
need to continue using a relational database, you may elect to use either
Amazon RDS (MySQL, Oracle, or SQL Server), or one of the many
preconfigured Amazon EC2 database AMIs. You are also free to create
your own Amazon EC2 instance, and install your own choice of database
software.
• Joins and/or complex transactions—While many solutions are able to
leverage Amazon DynamoDB to support their users, it’s possible that
your application may require joins, complex transactions, and other
relational infrastructure provided by traditional database platforms. If
this is the case, you may want to explore Amazon RDS or Amazon EC2
with a self-managed database.
• BLOB data—If you plan on storing large (greater than 64 KB) BLOB
data, such as digital video, images, or music, you’ll want to consider
Amazon S3. However, Amazon DynamoDB still has a role to play in this
scenario, for keeping track of metadata (e.g., item name, size, date
created, owner, location, and so on) about your binary objects.
• Large data with low I/O rate—Amazon DynamoDB uses SSD drives and
is optimized for workloads with a high I/O rate per GB stored. If you plan
to store very large amounts of data that are infrequently accessed, other
storage options, such as Amazon S3, may be a better choice.
Amazon ElastiCache
ElastiCache is a web service that makes it easy to deploy, operate, and scale a
distributed, in-memory cache in the cloud.32 ElastiCache improves the
performance of web applications by allowing you to retrieve information from a
fast, managed, in-memory caching system, instead of relying entirely on slower
disk-based databases. ElastiCache supports two popular open-source caching
engines: Memcached and Redis.
Page 38 of 50
• Memcached–ElastiCache is protocol-compliant with Memcached, a

widely adopted memory object caching system, so code, applications, and
popular tools that you use today with existing Memcached environments
will work seamlessly with the service.
• Redis–A popular open-source in-memory key-value store that supports

data structures such as sorted sets and lists. ElastiCache supports Redis
master / slave replication which can be used to achieve cross Availability
Zone redundancy.

ElastiCache improves application performance by storing critical pieces of data
in memory for low-latency access. It is frequently used as a database front end
in read-heavy applications, improving performance and reducing the load on
the database by caching the results of I/O-intensive queries. It is also frequently
used to manage web session data, to cache dynamically-generated web pages,
and to cache results of computationally-intensive calculations, such as the
output of recommendation engines. For applications that need more complex
data structures than strings, such as lists, sets, hashes, and sorted sets, the
Redis engine is often used as an in-memory NoSQL database.
Performance
Performance of a cache layer is very dependent on the caching strategy and the
hit rate at the application level, so it is difficult to provide general guidance.
Choose the total amount of cache memory needed based on the size of your
dataset and the expected access pattern. Then divide this by the memory per
cache node to get the required number of cache nodes. Make sure that you can
maintain acceptable performance without overloading the database backend in
the event of the failure and replacement of one or more cache nodes. You can
easily add or remove cache nodes from a running cluster, but you cannot change
the cache node type in a running cluster.

By definition, an in-memory cache stores transient data, or transient copies of
durable data, so data durability is managed elsewhere. It is still desirable to
have your cache be as fault-tolerant and highly-available as possible.
Page 39 of 50
With the Memcached engine, all ElastiCache nodes in a single cache cluster are
provisioned in a single Availability Zone. ElastiCache automatically monitors
the health of your cache nodes and replaces them in the event of network
partitioning, host hardware, or software failure. In the event of cache node
failure, the cluster remains available, but performance may be reduced due to
time needed to repopulate the cache in the new “cold” cache nodes. To provide
enhanced fault-tolerance for Availability Zone failures or cold-cache effects, you
can run redundant cache clusters in different Availability Zones.
With the Redis engine, ElastiCache supports replication to up to five read

replicas for scaling. To improve availability, you can place read replicas in other
Availability Zones. ElastiCache monitors the primary node, and if the node
becomes unavailable, ElastiCache will repair or replace the primary node if
possible, using the same DNS name. If the primary cache node recovery fails or
its Availability Zone is unavailable, you can failover from the primary node to
one of the read replicas with an API call.
Cost Model
With ElastiCache, you pay only for what you use and there is no minimum fee.
ElastiCache has only a single pricing component: pricing is per cache node-hour
consumed. For new customers, AWS provides a free usage tier that includes up
to 750 hours usage of a micro cache node. Pricing information can be found at
Amazon ElastiCache Pricing.33

ElastiCache is highly scalable and elastic. You can choose from a range of cache
node types supporting from 213 MB up to 68 GB of memory per node. You can
add or delete cache nodes to your cache cluster at any time to meet your
application load. Auto Discovery enables automatic discovery of Memcached
cache nodes by ElastiCache Clients when the nodes are added to or removed
from an ElastiCache cluster.
Interfaces
To control and manage ElastiCache—to create, describe, reboot, modify, and
destroy cache clusters—you can use the AWS Management Console, the
ElastiCache command line tools, the HTTP Query API, and various SDKs.
Page 40 of 50
To read and write data to ElastiCache cache cluster, you simply use the normal
Memcached and Redis APIs. Existing applications using Memcached or Redis
can use ElastiCache with almost no modifications other than changing the port
or DNS name used to connect. For a Memcached application, you use standard
operations like get, set, incr and decr in exactly the same way as you would in
your existing Memcached deployments. For a Redis application, you use GET,
SET, EXPIRE, and the various flavors of PUSH and POP exactly as you do with
existing Redis application.
ElastiCache for Memcached is protocol-compliant with Memcached, supporting

both the text and the binary protocols. As a result, you can switch to using
ElastiCache without recompiling or re-linking your applications–the libraries
you use will continue to work. Similarly, ElastiCache for Redis is protocol-
compliant with open source Redis. Code, applications, drivers and tools that
customers use today with their existing standalone Redis data store will
continue to work with ElastiCache for Redis.
Anti-Patterns
Amazon ElastiCache has the following anti-patterns:
• Persistent data—If you need very fast access to data, but also need
strong data durability (persistence), Amazon DynamoDB is probably a
better choice.
Amazon Redshift
Amazon Redshift is a fast, fully-managed, petabyte-scale data warehouse service
that makes it simple and cost-effective to efficiently analyze all your data using
your existing business intelligence tools.34 It is optimized for datasets that range
from a few hundred gigabytes to a petabyte or more.
Traditional data warehouses require significant time and resources to

administer, especially for large datasets. The financial cost associated with
building, maintaining, and growing self-managed, on-premises data warehouses
is very high. Amazon Redshift manages the work needed to set up, operate, and
scale a data warehouse, from provisioning the infrastructure capacity to
automating ongoing administrative tasks such as backups and patching.
Page 41 of 50

Amazon Redshift is ideal for analyzing large datasets using your existing
business intelligence tools. Organizations are using Amazon Redshift to do the
following:
• Analyze global sales data for multiple products
• Store historical stock trade data
• Analyze ad impressions and clicks
• Aggregate gaming data
• Analyze social trends
• Measure clinical quality, operation efficiency, and financial performance

in the health care space
Performance
Amazon Redshift uses a variety of innovations to obtain very high query
performance on datasets ranging in size from hundreds of gigabytes to a
petabyte or more. It uses columnar storage, data compression, and zone maps
to reduce the amount of I/O needed to perform queries. Amazon Redshift has a
massively parallel processing (MPP) architecture that parallelizes and
distributes SQL operations to take advantage of all available resources. The
underlying hardware is designed for high performance data processing that uses
local attached storage to maximize throughput.

Amazon Redshift has multiple features that enhance the reliability of your data
warehouse cluster. Amazon Redshift stores three copies of your data—all data
written to a node in your cluster is automatically replicated to other nodes
within the cluster, and all data is continuously backed up to Amazon S3.
Snapshots are automated, incremental, and continuous. Amazon Redshift stores
your snapshots for a user-defined period, which can be from one to thirty-five
days. At any time, you can create one or more manual snapshots, which are
retained until explicitly deleted. Amazon Redshift also continuously monitors
the health of the cluster and automatically re-replicates data from failed drives
and replaces nodes as necessary.
Page 42 of 50
Cost Model
With Amazon Redshift, you can pay as you go and there are no upfront costs.
Amazon Redshift has three pricing components: data warehouse node hours,
backup storage, and data transfer. Compute node hours are the total number of
hours run across all compute nodes for the billing period. Backup storage is the
storage associated with automated and manual snapshots for an Amazon
Redshift data warehouse cluster. Increasing the backup retention period or
taking additional snapshots increases the backup storage consumed by the
Amazon Redshift data warehouse cluster. There is no additional charge for
backup storage up to 100% of your provisioned storage for an active data
warehouse cluster.
There is no data transfer charge for data transferred to or from Amazon

Redshift outside of Amazon Virtual Private Cloud (Amazon VPC). Data transfer
to or from Amazon Redshift in Amazon VPC accrues standard AWS data
transfer charges.
Pricing information can be found at Amazon Redshift Pricing.35

Amazon Redshift provides “pushbutton scaling” of compute nodes within a data
warehouse cluster. With a few clicks of the AWS Management Console or a
simple API call, you can easily scale the number of nodes in your data
warehouse cluster up or down as your performance or capacity needs change.
An Amazon Redshift data warehouse cluster can be started with as little as a
single 2 TB XL node and scale all the way to a hundreds 16 TB 8XL nodes for 1.6
PB of compressed user data. Amazon Redshift will place your existing cluster
into read-only mode, provision a new cluster of your chosen size, and then copy
data from your old cluster to your new one in parallel. Queries can continue
running against the old cluster while the new one is being provisioned. Once the
data has been copied to the new cluster, Amazon Redshift will automatically
redirect queries to the new cluster and remove the old cluster.
Interfaces
The Amazon Redshift Query API provides a management interface to manage
data warehouse clusters programmatically. Additionally, the AWS SDKs for
Java, .NET, and other languages provide class libraries that wrap the underlying
Amazon Redshift API to simplify your programming tasks. If you prefer a more
Page 43 of 50
interactive way of managing clusters, you can use the Amazon Redshift console
and the AWS CLI.
The Amazon Redshift APIs do not provide a data interface. Amazon Redshift is
a SQL data warehouse and uses industry standard ODBC and JDBC connections
and PostgreSQL drivers. Once you’ve provisioned your cluster, you can connect
to it, start loading data, and run queries using the same SQL-based tools and
business intelligence applications you use today. For more information, see the
Amazon Redshift Partners page.36
Data can be loaded into Amazon Redshift from a range of data sources including
Amazon S3, Amazon DynamoDB, and AWS Data Pipeline. Amazon Redshift
attempts to load data in parallel into each compute node to maximize the rate at
which data can be ingested into the data warehouse cluster. For more
information on loading data into Amazon Redshift, see the Amazon Redshift
Getting Started Guide.37
Anti-Patterns
Amazon Redshift has the following anti-patterns:
• OLTP workloads—Amazon Redshift is a column-oriented database

suited to data warehouse and analytics, where queries are typically
performed over very large datasets. If your application involves online
transaction processing, a traditional row-based database system, such as
Amazon RDS, is a better match.
• BLOB data—If you plan on storing binary (e.g., video, pictures, or

music), you’ll want to consider Amazon S3.
Databases on Amazon EC2

Amazon EC2, together with Amazon EBS volumes, provides an ideal platform
for you to operate your own self-managed relational database in the cloud.38
Many leading database solutions are available as prebuilt, ready-to-use Amazon
EC2 AMIs, including IBM DB2 and Informix, Oracle Database, MySQL,
Microsoft SQL Server, PostgreSQL, Sybase, EnterpriseDB, and Vertica.
Page 44 of 50

Running a relational database on Amazon EC2 and Amazon EBS is the ideal
scenario for users whose application requires a specific traditional relational
database not supported by Amazon RDS, or for those users who require a
maximum level of administrative control and configurability.
Performance
The performance of a relational database instance on Amazon EC2 depends on
many factors, including the Amazon EC2 instance type, the number and
configuration of Amazon EBS volumes, the database software and its
configuration, and the application workload. In general, you can expect
database performance on Amazon EC2 to be similar to the performance of the
same database installed on similarly configured on-premises equipment. We
encourage you to benchmark your actual application on several Amazon EC2
instance types using several storage configurations to select the best
configuration.
To increase database performance you can scale up memory and compute

resources by choosing a larger Amazon EC2 instance size. For database storage,
it is usually best to use Amazon EBS Provisioned IOPS volumes. To scale up I/O
performance, you can increase the Provisioned IOPS, change the number of
Amazon EBS volumes, or use software RAID 0 (disk striping) across multiple
Amazon EBS volumes, which will aggregate total IOPS and bandwidth. In many
cases, you can also scale the total database system performance by scaling
horizontally with database clustering, replication, and multiple read slaves. In
general, you have the same database performance tuning options in the Amazon
EC2 environment that you have in a physical server environment.

Relational databases on Amazon EC2 provide persistent storage for structured
data using Amazon EBS volumes as the data store, so all the notes about the
durability and availability of Amazon EBS data apply here as well. And, again,
the basic durability and availability of relational data stored on Amazon EBS
volumes can be further enhanced by using Amazon EBS snapshots, or by using
third-party database backup utilities (such as Oracle’s RMAN) to store database
backups in Amazon S3.
Page 45 of 50
Cost Model
By running a database on Amazon EC2, you pay only for what you use and there
is no minimum fee. The cost of running your own database on Amazon EC2
depends on the size and number of Amazon EC2 instances used to run your
database, the size of the Amazon EBS volumes used for database storage, the
amount of data transferred in and out of Amazon EC2, and, in many cases, the
license cost of the third-party database software. Many open-source database
packages use a no-cost license model; some commercial software vendors use
the Amazon DevPay model; many others provide a bring-your-own-license
model. Contact your database software vendor or Amazon Web Services to
understand the license cost pricing model that applies. Pricing information for
Amazon EC2, Amazon EBS, and data transfer can be found at Amazon EC2
Pricing.39

In many cases, users of traditional relational database solutions on Amazon EC2
can take advantage of the scalability and elasticity of the underlying AWS
platform. For example, after you configure an Amazon EC2 instance with your
database solution, you can bundle the instance into a custom AMI, either by
using the bundle commands for instance store AMIs, or by using the create
image command for Amazon EBS AMIs. You can then create multiple new
instances of your database configuration in a few moments.
Anti-Patterns
Running your own relational database on Amazon EC2 is a great solution for
many users, but a number of scenarios exist where other solutions might be the
better choice. Self-managed relational databases on Amazon EC2 have the
following anti-patterns:
• Index and query-focused data—Many cloud-based solutions don’t

require advanced features found in a relational database, such as joins or
complex transactions. If your application is more oriented toward
indexing and querying data, you may find Amazon DynamoDB to be
more appropriate for your needs, and significantly easier to manage.
• Numerous BLOBs—Many relational databases support BLOBs (audio

files, videos, images, and so on). If your application makes heavy use of
Page 46 of 50
them, you may find Amazon S3 to be a better choice. You can use a
database to manage the metadata.
• Automatic scaling—Users of relational databases on AWS can, in many

cases, leverage the scalability and elasticity of the underlying AWS
platform, but this requires system administrators or DBAs to perform a
manual or scripted task. If you need pushbutton scaling or fully-
automated scaling, you may opt for another storage choice such as
Amazon DynamoDB or Amazon RDS.
• MySQL, Oracle, SQL Server—If you are running a self-managed

MySQL, Oracle, or SQL Server database on Amazon EC2, you should
consider the automated backup, patching, Provisioned IOPS, replication,
and pushbutton scaling features offered by a fully-managed Amazon RDS
database.
Contributors
The following individuals and organizations contributed to this document:
• Joseph Baron
• Sanjay Kotecha
Further Reading
For additional help, see the following sources.
Cloud Storage Use Cases

For illustrations of real-world usage of AWS storage options, see the companion
Storage Options in the AWS Cloud: Use Cases whitepaper.
AWS Storage Services

• Amazon S3
• Amazon Glacier
• Amazon EBS
Page 47 of 50
• Amazon EC2 Instance Store Volumes

(See sections on instance types, instance storage, and block device
mapping)
• AWS Import/Export
• AWS Storage Gateway
• Amazon CloudFront
• Amazon SQS
• Amazon RDS
• Amazon DynamoDB
• Amazon ElastiCache
• Amazon Redshift
Other Resources
• AWS SDKs, IDE Toolkits, and Command Line Tools
• Amazon Web Services Simple Monthly Calculator
• Amazon Web Services Blog
• Amazon Web Services Forums
• Running Databases on AWS
• Database AMIs in AWS Marketplace

(Search AWS Marketplace for “database”)
• Public Database AMIs

(Search public AMI list for “database” or for a particular databases)
• AWS Free Usage Tier
• Public Data Sets on AWS
• AWS Case Studies
Page 48 of 50
Notes
1 http://media.amazonwebservices.com/AWS_Storage_Use_Cases.pdf
2 https://aws.amazon.com/s3/
3 http://aws.amazon.com/free/
4 http://aws.amazon.com/s3/pricing/
5 http://aws.amazon.com/glacier/
6 http://aws.amazon.com/glacier/pricing/
7 http://aws.amazon.com/sns/
8 http://docs.aws.amazon.com/AmazonS3/latest/dev/object-lifecycle-
mgmt.html
9 http://aws.amazon.com/ebs/
10 http://aws.amazon.com/ec2/pricing/#EBS
11 http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/
12 Most Amazon EC2 instance types provide local instance storage volumes;
however, the micro and the M3 instance are Amazon EBS-only, and do not
provide local instance storage. Also, instances that use Amazon EBS for the
root device (boot from Amazon EBS) do not expose the instance store volumes
by default. If desired, you can expose the instance store volumes at instance
launch time by specifying a block device mapping. For more information, see
the Amazon EC2 User Guide.
13
http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/storage_instances.
html
14
http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/high_storage_insta
nces.html
15 http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-instance-
lifecycle.html
16 http://aws.amazon.com/ec2/pricing/
17 http://aws.amazon.com/importexport/
18 http://aws.amazon.com/importexport/pricing/
Page 49 of 50
19 http://aws.amazon.com/storagegateway/
20 http://docs.amazonwebservices.com/storagegateway/latest/userguide/
21 http://aws.amazon.com/directconnect/
22 http://aws.amazon.com/storagegateway/pricing/
23 https://console.aws.amazon.com/storagegateway
24 http://aws.amazon.com/cloudfront/
25 http://aws.amazon.com/cloudfront/pricing/
26 http://aws.amazon.com/sqs/
27 http://aws.amazon.com/sqs/pricing/
28 http://aws.amazon.com/rds/
29 http://aws.amazon.com/rds/pricing/
30 http://aws.amazon.com/dynamodb/
31 http://aws.amazon.com/dynamodb/pricing/
32 http://aws.amazon.com/elasticache/
33 http://aws.amazon.com/elasticache/pricing/
34 http://aws.amazon.com/redshift/
35 http://aws.amazon.com/redshift/pricing/
36 http://aws.amazon.com/redshift/partners/
37 http://docs.aws.amazon.com/redshift/latest/gsg/welcome.html
38 http://aws.amazon.com/ec2/
39 http://aws.amazon.com/ec2/pricing/
Page 50 of 50

Aws Storage Options

Uploaded by

Copyright:

Available Formats

Aws Storage Options

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Aws Storage Options

Uploaded by

Copyright:

Available Formats

Storage Options in the

In a separate companion whitepaper, we present several storage use cases that

Traditional vs. Cloud-Based Storage Alternatives

• Memory—In-memory storage, such as file caches, object caches, in-

• Message Queues—Temporary durable storage for data sent

• Storage area network (SAN)—Block devices (virtual disk LUNs) on

• Network attached storage (NAS)—NAS storage provides a file-level

• Databases—Structured data is typically stored in some kind of database,

database, or a data warehouse. The underlying database storage typically

• Backup and Archive—Data retained for backup and archival purposes is

Each of these traditional storage options differs in performance, durability, and

As we’ll see throughout this whitepaper, AWS offers multiple cloud-based

This whitepaper examines the following AWS cloud storage options:

Amazon S3 Scalable storage in the cloud

Amazon Glacier Low-cost archive storage in the cloud

Persistent block storage volumes for Amazon EC2 virtual

AWS Import/Export Large volume data transfer

AWS Storage Gateway Integrates on-premises IT environments with cloud storage

Amazon CloudFront Global content delivery network (CDN)

Amazon SQS Message queue service

Managed relational database server for MySQL, Oracle, and

Amazon DynamoDB Fast, predictable, highly-scalable NoSQL data store

Amazon ElastiCache In-memory caching service

Fast, powerful, full-managed, petabyte-scale data warehouse

Amazon Simple Storage Service (Amazon S3)

Ideal Usage Patterns

Amazon S3 is also frequently used to host entire static websites. Amazon S3

Finally, Amazon S3 is often used as a highly durable, scalable, and secure

To speed access to relevant data, many developers pair Amazon S3 with a

Durability and Availability

Multi-Factor Authentication (MFA) Delete. With this option enabled on a

5 GB of Amazon S3 storage.3 Pricing information can be found at Amazon S3

Scalability and Elasticity

Most developers building applications on Amazon S3 use a higher-level toolkit

• File system—Amazon S3 uses a flat namespace and isn’t meant to serve

• Structured data with query—Amazon S3 doesn’t offer query

• Rapidly changing data—Data that must be updated very frequently

• Backup and archival storage—Data that requires long-term encrypted

• Dynamic website hosting—While Amazon S3 is ideal for websites with

Ideal Usage Patterns

Durability and Availability

Pricing information can be found at Amazo n Glacier P ricing.6

Scalability and Elasticity

First, Amazon Glacier provides a native, standards-based REST web services

Second, Amazon Glacier can be used as a storage class in Amazon S3 by using

• Rapidly changing data—Data that must be updated very frequently

• Real time access—Data stored in Amazon Glacier is not available in real

Amazon Elastic Block Store (Amazon EBS)

Amazon EBS also provides the ability to create point-in-time snapshots of

Ideal Usage Patterns

Provisioned IOPS volumes are designed to deliver predictable, high performance

Alternatively, you could stripe your data across multiple similarly-provisioned

Durability and Availability

consistency, it is a best practice to briefly pause any writes to the volume, or

Scalability and Elasticity

1. Detach the original Amazon EBS volume.

• Highly-durable storage—If you need very highly-durable storage, use

Amazon EC2 Instance Store Volumes

In addition, the storage-optimized Amazon EC2 instance family provides special-