MX Player: Achieving Massive Scalability Using Amazon EKS with Spot Instances

Опубликовано 20 декабря 2022, 19:59

MX Player is one of the largest AVOD based OTT platform in India with Billion + downloads and a massive MAU, during COVID times the platform saw a massive jump in userbase and similar kind of increase in content produced and ingested into the platform. To deliver the content to this enormous userbase platform in a seamless manner, the other challenge was to cater to different types of devices with all shapes and forms ranging from mobile, tablets to TVs, to support this plethora of devices the platform needs to transcode raw ingested data into different types of profiles, not only the data needs to be transcoded to support varying network characteristics given the users are located in Tier 1, 2, and 3 cities where network bandwidths vary significantly. Using EKS with Spot based on accelerated compute instances like G4DN, MXP was able to achieve this massive scalability in transcoding without compromising the quality with substantial amount of savings in terms of time and cost. Another aspect is the efficient and secure content delivery to the end user, for this MXP leverages AWS Cloudfront with a combination of AWS WAF and AWS Shield.

Check out more resources for architecting in the #AWS​​​ cloud:
amzn.to/2ZIbygO

#AWS #AmazonWebServices #CloudComputing #ThisIsMyArchitecture