Slides for this talk: barcamp_bangalore_2013_pavan_verma
Amazon Web Services (AWS) provides a global-scale elastic computing infrastructure for hosting applications. It is the reason why companies such as Dropbox have been able to scale up so quickly.
Even though AWS datacenters are better maintained than the average datacenter, they do suffer from outages. However well a datacenter is designed, failures do happen! Hardware and software does crash! AWS suffered a big outage recently on 24-Dec-2012 in its biggest datacenter. The story that people could not watch their favorite movies on Christmas eve using Netfix streaming is well known.
In this session, we will discuss the different types of failures that can happen in a datacenter and how you can architect your application deployment to safeguard against these failures resulting in better availability of your application, happier customers and a better reputation. We will discuss failures ranging from single component failures such as hard drive or a server to disasters that can cause an entire datacenter to fail.