MediaMath Maximizes Data Access and Minimizes Latency on AWS

MediaMath is an advertising technology company that helps advertisers and agencies efficiently run media campaigns. The company analyzes 120 billion advertising opportunities per day, with throughput over 2 million bid requests per second. MediaMath processes 3 TB of data per day on the Amazon Web Services (AWS) Cloud, collecting log-level data on real-time bidding wins, web events such as page visits, product purchases, and other offline data.

In this re:Invent 2014 session, learn how MediaMath migrated from a legacy batch processing architecture to a real-time streaming data platform on AWS for low-latency consumer analysis, advertising analytics, bid decisioning, and optimization. MediaMath used Amazon Kinesis, Amazon Elastic Compute Cloud (Amazon EC2) Spot Instances, Amazon EMR, and Amazon Simple Storage Service (Amazon S3) to create a terabyte-scale data streaming architecture with real-time access, record-by-record routing and processing, and multiple consumers. As a result, the company was able to dramatically expand access to low-latency data for internal teams and external partners.

To learn more, watch the video below and visit aws.amazon.com/advertising-marketing.