A genomics research institute runs high-performance DNA sequencing data analysis that requires 128 compute nodes for exactly 6 hours every Friday. For the rest of the week, the institute only needs 2 compute nodes to allow researchers to access historical data. Which of the following best describes the cloud economics benefit of running this workload on AWS rather than on-premises?
- The institute can leverage elasticity to pay for the 128 compute nodes only during the 6 hours of active analysis, avoiding the costs of idle on-premises infrastructure.Answer
- BThe institute can transition from a variable operational expense (OpEx) model to a fixed capital expense (CapEx) model, securing long-term predictable compute costs.
- CThe institute can purchase 128 Standard Reserved Instances to cover the Friday peak, ensuring that the necessary compute capacity is reserved and paid for throughout the entire week.
- DThe institute can rearchitect the workload into a single, tightly coupled monolithic application to reduce the network communication costs between separate nodes.
Answer
The institute can leverage elasticity to pay for the 128 compute nodes only during the 6 hours of active analysis, avoiding the costs of idle on-premises infrastructure.
The correct answer correctly identifies elasticity and the pay-as-you-go billing model. In an on-premises deployment, the institute would have to purchase and maintain 128 servers to handle the Friday workload, meaning 126 servers would sit idle and waste resources for the rest of the week. With AWS, the institute can scale up to 128 nodes for the 6 hours of processing and then scale back down to 2 nodes, paying only for the resources they consume.
Step-by-Step Solution
Key Concept
Cloud elasticity and pay-as-you-go pricing model vs. on-premises over-provisioning
Estimated Time:1m 30s