A logistics company is designing an AWS architecture for two distinct workloads. The first workload is a containerized microservices API for processing real-time order transactions; it experiences highly variable traffic, and the company wants to run it without managing the underlying servers. The second workload is an on-demand batch analytics job that processes historical shipping records, takes approximately 4 hours to complete, requires a customized Linux environment with proprietary software, and can be paused or restarted without impact if interrupted. Which two of the following compute options should the company select to meet these requirements? (Select two.)
- AWS Fargate to deploy the containerized microservices API without managing EC2 instancesCevap
- Amazon EC2 Spot Instances to run the batch analytics jobCevap
- CAWS Lambda to run the batch analytics job as a serverless function
- DAmazon EC2 Reserved Instances to run the batch analytics job
Cevap
The correct answers are the option to use AWS Fargate to deploy the containerized microservices API without managing EC2 instances, and the option to use Amazon EC2 Spot Instances to run the batch analytics job.
AWS Fargate is the ideal serverless compute option for running containers (ECS/EKS) when developers want to eliminate the operational overhead of provisioning, configuring, and scaling virtual machines. Amazon EC2 Spot Instances offer the most cost-effective pricing model for workloads that are tolerant of interruptions (such as batch data processing) and require custom operating systems or applications running beyond serverless time limits.
Adım Adım Çözüm
Anahtar Kavram
Selecting the appropriate AWS compute service and purchasing model based on execution time limits, server management preferences, containerization requirements, and fault tolerance.