A company needs to design a solution to process videos that users upload to an Amazon S3 bucket. Each video file is approximately 1 GB in size and takes approximately 20 minutes to process. During peak hours, the company expects to process approximately 100 simultaneous uploads. The video file processing is stateless and can run in parallel as soon as the video files arrive in the S3 bucket.
Which solution will meet these requirements in the MOST cost-effective way?
The workload isstateless,parallel, and each job runs for about20 minutes, which makes Lambda a poor fit because Lambda has a maximum execution time of 15 minutes. AWS Fargate is a strong match for event-driven container workloads because it removes server management overhead, andFargate Spotlowers cost for interruption-tolerant processing. AWS also provides patterns for runningevent-driven workloads at scale with Fargate, and EventBridge can be used to trigger container tasks from S3-originated events. Compared with EKS or EC2 On-Demand, ECS on Fargate Spot is the most cost-effective managed container approach here.
============
Currently there are no comments in this discussion, be the first to comment!