Engineering / December 19, 2016

Servers on Fleet : Elasticity at Scale

At Eyeview, we build systems that need to handle hundreds of thousands of requests per second. Amazon Web Services provides most of the tools and services you need to create such elastic and scalable applications. However, like most technologies, it has quite a few tradeoffs that you need to account for including cost, flexibility, and best practices.

In May, we conducted a session on the AWS as part of the AWS New York Meetup group. Our presentation focused on the evolution of our real-time bidding system. We discussed how we started from an elastic — though naive — service from day one. We also went over how we moved from a costly architecture to a cost-efficient structure that is using different Amazon Web Services, such as elastic load-balancing, spot fleets, and an availability-zone-contained architecture.

This session will give you some tips from from our trenches about the little details that matter when creating a highly scalable application using different Amazon services.

 
Shahar Cizer Kobrinksy

Shahar Cizer Kobrinksy VP of Architecture & Scale

Date: 12.19.2016
Tags: