Amazon CloudFront Scales Environment For Super Bowl Traffic
Samsung wanted to create a new type of advertising campaign for customers that connected on their Smart TVs during the Super Bowl during Katy Perry’s Halftime Show. The challenge at hand was – how do you create and maintain more than 4,000,000 concurrent connections and all of that data being transferred in and out of Samsung Smart TVs during the biggest, live television event in history?
Handling Massive Traffic & Massive Data
Onica worked diligently to make sure Samsung’s environment could scale and handle the Super Bowl-sized traffic load all at once. Since the Halftime Show is such a highly watched, live event, the tremendous flash of traffic arrives and then disappears much quicker than normal television content. The team at Onica partnered with Samsung’s developers to model this traffic load in the time building up to the big game. This ambitious project not only required developers to write the code needed for Samsung’s Smart TVs, but also infrastructure specialists that could help to build a scalable and elastic infrastructure that could handle 4,000,000 concurrent connections.
A Super Bowl Sized Solution
Samsung uses Amazon CloudFront, Amazon’s Content Delivery Network (CDN) to push data back from their Smart TVs into AWS. The Smart TVs have a logging system that registers each box within AWS. Based on this information, Samsung can then tailor content and a unique user experience to each Smart TV viewer through Amazon CloudFront.
Onica and Samsung tested and built the Super Bowl Halftime Show solution for 500,000 total requests per second, all being driven through Amazon CloudFront. Onica performed a significant amount of load tests on a number of occasions, with each one increasing up to the event. To test durability, Onica tried to break the system and was successful in making 1,000,000 concurrent requests per second.
In order to perform these load tests, AeonX used a variety of tools. The Samsung team also provided Live TVs across America to help simulate these load tests, which created a realistic experience for what the Super Bowl load might look like during the Halftime Show.