Open In App

The Engine Behind Disney + Hotstar: A Peek into Their Backend Infrastructure

Last Updated : 18 Nov, 2023
Improve
Improve
Like Article
Like
Save
Share
Report

On 15 November 2023, when India was playing against New Zealand at the World Cup semifinal match in Wankhede Stadium in Mumbai more than 5 crore people were watching the match on the Disney+ Hotstar OTT(over-the-top) apps on a global level. This number had broken the streaming platform’s own concurrency record of 3.5 crore viewers, which was created during the India vs Pakistan match earlier in the tournament. We will discuss how the Disney + Hotstar app is able to ensure such powerful scalability, on a consistent basis.

Key Elements in their Backend System:

Disney+ Hotstar relies on a robust backend system to ensure seamless streaming. The key elements in their backend system:

Hosting with AWS (Amazon Web Service):

AWS serves as the primary hosting provider for Disney+ Hotstar. It relies on AWS (Amazon Web Services) as the backbone of their hosting, and it provides details about the infrastructure, including the use of AWS instances during high-impact events. It offers a scalable and reliable cloud computing platform, and allow Disney+ Hotstar to efficiently manage their infrastructure based on demand.

CDN (Content Delivery Network) Partner:

Disney+ Hotstar CDN partner is Akamai. It plays important role in delivering content efficiently to users. It optimize the media files for delivering, by ensuring low latency and high performance during streaming.

Data Store:

Disney+ Hotstar employ Amazon S3 Storage service for storing data. S3 Storage is an object storage services offered by AWS. Amazon S3 provides Disney+ Hotstar with a scalable, cost-effective, and feature-rich solution for storing and delivering the vast amount of media content associated with a streaming service. The flexibility and integration capabilities of Amazon S3 contribute to the overall efficiency and reliability of Disney+ Hotstar’s backend infrastructure.

Note:

Disney+ Hotstar, 100% traffic is supported by EC2 and S3 object storage for data storing. For EC2, Disney+ Hotstar can choose instances with the right balance of CPU, memory, storage, and networking capabilities to meet their specific requirements.

Instance Mix :

Instance Mix refers to the combination of different computing instances that are utilized within a system or application. In Disney+ Hostar, Instance Mix means as a part of their backend infrastructure. There are two types of instance used in Disney+ Hotstar:

  • On-Demand Instances: On-Demand instances are virtual servers that are provisioned and run on a pay-as-you-go basis. These instances are suitable for regular workloads and typically used when there is a consistent need for computing resource.
  • Spot Instances: Spot instances are cost-effective option for acquiring computing capacity. These instances allow users to bid for unused EC2 capacity at potentially lower prices than on-demand instances. Spot instances are suitable for workloads that can be interrupted and don’t require continuous availability.

The use of an Instance Mix suggests Disney+ Hotstar employs a combination of both on-demand and spot instances in their infrastructure. This approach offers a balance between cost efficiency and performance.

Smart Cost Management with Machine Learning and Data Analytics:

Disney+ Hotstar can achieve a more adaptive and responsive infrastructure. These technologies enable the system to learn from historical data, make predictions, and dynamically optimize resource usage, ultimately leading to cost savings while maintaining a high level of performance during varying workloads.

Data Processing :

Disney+ Hotstar processes massive amounts of data daily using AWS EMR (Elastic MapReduce) clusters. EMR is a cloud-based big data platform that enables processing of large datasets using popular frameworks such as Apache Spark and Apache Hadoop.

Scaling with Ease (Infrastructure Set up):

To handle peak loads during high impact event like World Cup matches, Disney+ Hotstar scales its infrastructure with:

500+ AWS CPU instances (C4.4X Large or C4.8X Large) at 75% utilization.
16 TBs of RAM
8000 CPU cores
A peak speed of 32Gbps for data transfer

  • Disney+ Hotstar’s infrastructure during high-impact events includes over 500 AWS CPU instances, specifically from the C4 family with sizes C4.4X Large or C4.8X Large. These instances are operated at 75% utilization, demonstrating a strategic approach to balancing resource efficiency and scalability to handle the surge in user activity during popular events.
  • It allocates a significant amount of RAM (16 terabytes) to their infrastructure, underscoring the importance of robust memory resources in supporting the demands of streaming services.
  • A substantial number of CPU cores (8000) to their infrastructure, emphasizing the importance of computational power for efficiently handling the processing demands of a large-scale streaming service.
  • It’s infrastructure is designed to achieve a peak data transfer speed of 32Gbps, showcasing a commitment to providing a seamless and high-quality streaming experience.

Disney+ Hotstar uses these Andriod components for having a powerful infrastructure:

  • ViewModel for network communication.
  • Room for data persistence.
  • LifecycleObserver.
  • RxJava 2.
  • Dagger 2 and Dagger Android.
  • AutoValue.
  • Glide 4.
  • Gson.
  • Retrofit 2 + okhttp 3.
  • Chuck Interceptor for easy debugging.

Scalability Design:

The infrastructure is designed for scalability, ensuring that millions of users can access live streaming simultaneously without compromising performance.

Conclusion:

In summary, Disney+ Hotstar’s backend infrastructure is a well-thought-out combination of cloud services, content delivery optimization, data storage, cost-effective instance management, and scalability features. The use of advanced technologies and strategic design choices contributes to a reliable and efficient streaming platform for a massive user base.


Like Article
Suggest improvement
Share your thoughts in the comments

Similar Reads