Peiris explains that investing in Amazon Web Services (AWS) has supported the execution of his company’s vision in two ways: “It increased our agility by reducing our time to market and reduced our fixed cost base.”

He goes on to describe the company’s initial use of AWS: “We ran a nine-month project to transition all customer-facing websites to AWS to address all our infrastructure needs. The project was delivered in an agile manner, with the solution incorporating the latest pieces of AWS technology—including

“The project involved transitioning 10 quad core blade web servers and two 1-TB MySQL databases on the backend to 12 web servers (across three Availability Zones), one large Amazon RDS instance, a small number of memcached instances, and 10 ELBs.”

The following tools were used to develop Holiday Extras’ solution: Bash, Phing, PHP, AWS Java tools, TeamCity, SVN, Zabbix (for monitoring), and Patch (for deployment).

Peiris offers the following recommendations for other developers:

• Understand the cloud requires differing operational patterns (such as disposable technology and build for down time).
• Understanding how the variety of AWS tools work makes architecture design and implementation easier.
• Don’t expect to be able to complete a move in a very short time frame.
• Planning for outages is an important step.
• Have a look at some case studies as they provide valuable insight from people that have already gone through the process.
• Ideally, move small parts of your systems over one by one. This makes it easier to spot where problems might be occurring.
• Build a management console of your own that provides you with the functionality you’re looking to achieve and provides insight into your current infrastructure.
• Try to forget about looking after individual boxes in clusters, etc. If one is having difficulties for strange reasons, get rid of it and start a new instance.