An educational technology company is planning to deploy two distinct workloads on Amazon EC2:
1. A production student portal application that must run continuously and exhibits stable, predictable usage throughout the school year.
2. A machine learning model training job that runs periodically, can be safely paused and resumed, and needs the absolute lowest cost compute options.
Which two Amazon EC2 pricing models should the company select to run these workloads in the most cost-effective manner?
- Reserved Instances for the continuous student portal applicationCevap
- Spot Instances for the periodic machine learning training jobCevap
- COn-Demand Instances for the continuous student portal application
- DDedicated Hosts for the periodic machine learning training job
- EAWS Free Tier to run both workloads
Cevap
Reserved Instances for the continuous student portal application and Spot Instances for the periodic machine learning training job
For the continuous and predictable student portal application, Reserved Instances provide a significant discount (up to ) in exchange for a one-year or three-year commitment. For the periodic and interruptible machine learning model training job, Spot Instances offer the lowest cost (up to a discount) because they utilize unused EC2 capacity and can be reclaimed by AWS with a two-minute warning.
Adım Adım Çözüm
Anahtar Kavram
AWS EC2 pricing models selection based on workload requirements