Amazon EC2 to S3: A guide to AWS products

The AWS product list is as vast as it is confusing. Here's what you need to know about the most vital AWS products, up to and including Amazon EC2, along with pricing information.

AWS offers a multitude of cloud computing offerings that are both cost-effective and flexible. Whether you're looking for storage, computing capacity or online databases, it's easy to find cloud services that can soothe your IT needs.

There are so many AWS products, however, it can be overwhelming. To make it a little easier, we've compiled a list of some of the most commonly used services and their basic pricing information. Check out Amazon's website for a more detailed look at AWS products.

Amazon Elastic Compute Cloud (Amazon EC2)

Amazon EC2 offers cloud-based and resizable compute capacity. Amazon touts EC2 as making web-scale computing easier for developers. It enables quick scaling by offering developers a quick means to boot new server instances and allows users to maintain complete control of computing resources.

Pricing: New AWS customers can use EC2 free for up to 750 hours per month for one year. It also offers on-demand instances charged for compute capacity by hour or second, with a minimum of 60 seconds. For spot instances, Amazon charges for the specific time the instances are running, with pricing based on long-term supply-and-demand trends. Amazon also offers dedicated physical EC2 servers and reserved instances for applications with predictable usage.

Amazon Elastic Container Service (Amazon ECS)

This scalable container orchestration service from Amazon supports Docker containers and enables users to run and scale containerized applications on AWS. Amazon ECS essentially doubles as container orchestration software, managing and scaling virtual machine clusters and container scheduling.

Pricing: Amazon divides ECS into two pricing structures: Fargate Launch Type and EC2 Launch Type. Fargate is priced according to the virtual CPU and memory resources the user's containerized applications request, rounded up to nearest second from the time the container is pulled until the task terminates, with a minimum charge of one minute.

With EC2, there is no additional charge. Amazon only charges users for the AWS resources they create and store in running the application.

Amazon Relational Database Service (Amazon RDS)

Amazon RDS provides scalable relational databases in the cloud through a web service. Users can choose from a handful of database engines, including Amazon Aurora, PostgreSQL, MySQL, MariaDB, Oracle Database and SQL Server. Amazon RDS is also optimized for performance, I/O or memory, depending on user needs. Through AWS Database Migration Service, users can replicate and migrate existing databases. It can also automatically patch database software and perform regularly scheduled backups.

Pricing: Amazon RDS pricing is segmented into on-demand and reserved instances, with pricing dependent on relational database use cases. These include AWS products like Amazon Aurora, as well as Amazon RDS for MySQL, PostgreSQL, MariaDB, Oracle and SQL Server.

Amazon Simple Storage Service (Amazon S3)

Amazon S3 is an online storage service that offers unlimited data storage space in the cloud. Its web-service interface stores and retrieves data from anywhere online.

Pricing: New users will receive, for free, 5 GB of S3 storage, 20,000 Get Requests, 2,000 Put Requests, 15 GB of data transfer in and 15 GB of transfer out each month for one year.

Amazon Aurora

This managed relational database is compatible with MySQL and PostgreSQL. According to Amazon, it's five times and three times faster than MySQL and PostgreSQL databases, respectively. Amazon touts its security and reliability as equivalent to commercial databases, but at one-tenth the typical cost. Automation features include hardware provisioning, database setup, patching and backups, through Amazon Relational Database Service management. Amazon Aurora is also self-healing, with automatic scans and data recovery from its three AWS Availability Zones, where it replicates 10 GB data chunks six ways. To that end, Amazon Aurora storage starts from 10 GB, scaling to a max 64 TB.

Pricing: As with many AWS products, Aurora pricing is "pay as you go with no upfront fees," according to Amazon. Database instance, database storage and I/Os, backup storage, backtrack and data transfer rates vary according to region and MySQL and PostgreSQL editions.

Amazon CloudFront

CloudFront is a content delivery service that integrates with Amazon's other cloud services to provide an easy way for businesses and developers to distribute data through high-speed transfers. Optimized to work with EC2, Elastic Load Balancing, AWS Shield for dealing with DDoS and S3, CloudFront delivers static and streaming content using a global network of edge locations.

Pricing: It's pay-as-you-go. Prices start at $0.085 per gigabyte transferred out for the first 10 TB per month, depending on location. Reserved capacity pricing is available upon request.

Amazon CloudWatch

CloudWatch enables monitoring for EC2 and other Amazon cloud services, including Amazon S3, ECS, DynamoDB and AWS Lambda. For example, users can choose the EC2 instances they want to watch, and CloudWatch will gather and monitor the easily accessible data and analytics. Some of the metrics that it provides visibility into include CPU utilization, disk reads/writes and network traffic.

Pricing: Amazon breaks CloudWatch pricing into two tiers: free and paid. AWS products like EC2, Kinesis, Amazon S3 and others automatically send metrics to CloudWatch. However, there are limits to metrics, analytics, dashboards, alarms and logs, and it excludes custom events. Beyond that, users move into the pay-as-you-use paid tier.

Amazon DynamoDB

This managed document and key-value database offering includes security, in-memory caching, backup and restore, in addition to multi-region and multi-master support. The most significant value Amazon DynamoDB brings to enterprises comes from its scaling and reliability, providing single-digit millisecond response times to potentially more than 10 trillion requests per day and 20 million requests per second, according to Amazon.

Pricing: Pricing is tier-based. Amazon charges for application data reads/writes, while the user does not need to specify expected throughput, with DynamoDB's on-demand capacity mode. Through DynamoDB's provisioned capacity mode, Amazon charges reads/writes per second, based on the user's expected requirements. Through autoscaling, users can adjust capacity.

Amazon Elastic Block Store (Amazon EBS)

EBS provides Amazon cloud users with scalable block storage within AWS Cloud for AWS EC2 instances. As it does with many AWS products, Amazon touts its durability and availability, with each Amazon EBS volume automatically replicating with its Availability Zone.

Pricing: Amazon divides EBS pricing into tiers. The free tier includes 30 GB of storage, 2 million I/Os and 1 GB of snapshot storage. Amazon EBS Volumes pricing varies by region and volume type, including general purpose SSD, provisioned IOPS SSD, throughput optimized HDD, cold HDD and EBS snapshots to Amazon S3.

AWS Identity and Access Management (AWS IAM)

The AWS IAM offering enables AWS products platform customers to manage access to compute, storage, database and app services in AWS cloud, with access control concepts organized by users, groups and permissions applied to individual API calls. IAM supports external identity systems and platforms, including Microsoft Active Directory, which users can integrate.

Pricing: IAM is a feature of AWS, which Amazon offers free of charge as part of the account.

AWS Lambda

AWS Lambda is an event-driven cloud compute service that runs customer code for apps and back-end services handling administration tasks, such as server provisioning and management. This includes scaling, and, according to Amazon, Lambda precisely scales the application by running the code according to individual triggers.

Pricing: Amazon charges per 100 milliseconds of code execution and the number of trigger requests.

Amazon Route 53

Amazon Route 53 enables developers to translate URLs into numeric IP addresses, routing end users to internet applications. This cloud domain name system (DNS) web service is scalable and IPv6 compliant. Amazon Route 53 Traffic Flow assists with routing by enabling various routing types, including geo-DNS, geo-proximity, weighted round robin and latency-based. Amazon Route 53 also includes DNS server health check features and domain name purchasing and management.

Pricing: Amazon offers pay-as-you-go pricing for Route 53, including a monthly charge for each hosted zone, a charge per answered DNS query (with exceptions) and a monthly charge for each domain transferred into or registered through Route 53.

Amazon Virtual Private Cloud (Amazon VPC)

Through Amazon VPC, users can provision a logically isolated AWS Cloud section. Here, users can deploy AWS resources in a user-defined virtual private network. Users have total control over the environment, and this includes subnet creation, route table and network gateway configuration, and IP address range selection. Amazon VPC supports IPv4 and IPv6.

Pricing: Amazon VPC pricing is structured into three tiers: AWS Site-to-Site, AWS PrivateLink and NAT Gateway. Site-to-Site pricing starts at $0.05 per connection hour ($0.048 for Tokyo AWS Region and Osaka-Local AWS regions). PrivateLink pricing starts at $0.01 per hour and $0.01 per gigabyte data processed, with numbers depending on region. Amazon bases NAT Gateway pricing on each hour the NAT gateway is provisioned and available, starting at $0.045 per hour, and, as the amount of data processed, also starting at $0.045 per gigabyte, depending on region.

Elastic Load Balancing

Elastic Load Balancing is a networking service that automatically spreads out incoming application traffic across AWS EC2 instances, IP addresses and Lambda functions. Customers have the option of enabling it within a single Availability Zone or throughout multiple zones. The three load balancer types AWS offers all feature security, scaling and reliable availability. The service can also detect unhealthy instances and reroute traffic to healthy ones.

Pricing: AWS offers three Elastic Load Balancing pricing tiers: Application, Networking and Classic. With Application Load Balancer and Network Load Balancer, users are charged each hour or partial hour it is running, as well as for the Load Balancer Capacity Units used each hour. Amazon charges Classic Load Balancer users for each hour or partial hour of use, as well as each gigabyte of data transferred through.

Dig Deeper on AWS infrastructure

App Architecture
Cloud Computing
Software Quality