Moving Hadoop to the Cloud

Chapter 2. Overview and Comparison of Cloud Providers

This short chapter provides a brief background and history of the three major public cloud providers that are covered in this book. If you aren’t sure which one to use, this information may help you decide.

Amazon Web Services

Amazon Web Services (AWS) is, at the time of this writing, perhaps the dominant public cloud provider. It may be surprising to learn that its earliest services were launched in 2006, well before cloud computing grew popular:

• Elastic Compute Cloud (EC2), a service for provisioning computing resources on demand

• Simple Storage Service (S3), online storage for opaque data

The original primary intent of AWS was to resolve scalability issues within Amazon itself, but even in its initial proposal in 2003 it was recognized that the new infrastructure could also be offered to customers.

The next prominent services to be added to the AWS suite included:

• Elastic Block Store (EBS), persistent disk-like storage for EC2 instances, in 2008

• Elastic MapReduce (EMR), a service providing Hadoop-like clusters for running MapReduce (and later Apache Hive and Apache Pig) jobs, in 2009

• Relational Database Service (RDS), a service for managing relational database server instances running in AWS, also in 2009

In 2012, new AWS services focused on data storage; in that year alone the Redshift massive-scale data warehouse, the DynamoDB NoSQL database, and the Glacier data archival service were introduced. More recent ...