Business Case: Flotrak.com is a iOS based mobile app which allows users to share music clipps. Application allows users to select tracks on their mobile device and share a clipping with other users in the social network. Each clipping is approximately 1.8 mb in size and needs high processing capacity to handle peak loads. Expected download of the application remains over 1 million with a simultaneous play of about 200 thousand to 300 thousand users.
Solution Approach: Expected load ruled that the current Lamp Stack being used to run the services was not geared to take the expected peak load. It would have resulted in thousands of server being scaled thus increasing the cost of hosting manifold. TechMinfy provided an architecture which is highly scalable and available to address these issues. Revolving around a robust and Highly Available AWS architecture the app is now ready to face live usage of millions of users across the world with minimum latency and negligible downtime.
Previously Flotrak was using AWS RDS (MySQL) and TechMinfy migrated it to Aurora DB by creating a snap-shot of the existing database in MySQL and using the Migrate Snapshot utility successfuly migrated to Amazon Aurora. This project was completed in just 2 days and successfully delivered.
- Service developed using core LAMP stack
- Music file upload and download being managed via apache server
- Over a million users expected
- 200 thousand to 300 thousand simultaneous users expected
- Apache, PHP, MySQL,Route53, ELB, EC2, Cloudfront, S3, Elasticache, DynamoDb, Aurora DB and RDS
- Service data in increased to 15 records per second
- Music download and upload load removed from core app server
- Endless scale provided to media access
- Expected app server scaling reduced to 10-20 from 2000+