Which solution will meet these requirements?
Create an Auto Scaling group of m5.large Amazon EC2 Spot Instances behind an Application Load Balancer. Use an Amazon ElastlCache for Redis cluster to maintain the leaderboard.
Create an Auto Scaling group of c5.large Amazon EC2 Spot Instances behind an Application Load Balancer. Use an Amazon OpenSearch Service cluster to maintain the leaderboard.
Create an Auto Scaling group of c5.large Amazon EC2 On-Demand Instances behind an Application Load Balancer. Use an Amazon ElastiCache for Redis cluster to maintain the leaderboard.
Create an Auto Scaling group of m5.large Amazon EC2 On-Demand Instances behind an Application Load Balancer. Use an Amazon DynamoDB table to maintain the leaderboard.
Explanations:
While m5.large instances are suitable for general-purpose workloads, they are not optimized for high-performance computing (HPC). Additionally, using Spot Instances may lead to interruptions during critical gameplay, impacting performance. ElastiCache for Redis is appropriate for maintaining a leaderboard, but the instance type is suboptimal for HPC.
c5.large instances are optimized for compute-intensive tasks, which is beneficial for HPC. However, using Amazon OpenSearch Service is not suitable for maintaining a frequently changing leaderboard, as it is primarily designed for search and analytics rather than in-memory data structures like Redis.
c5.large instances are optimized for compute-intensive applications, making them ideal for HPC tasks. Using On-Demand Instances ensures reliability and reduces the risk of interruption. ElastiCache for Redis is well-suited for maintaining a leaderboard due to its fast in-memory data access and support for high throughput and low latency.
While m5.large instances could handle general workloads, they are not optimal for HPC. On-Demand Instances provide reliability, but using DynamoDB for maintaining a leaderboard is less efficient compared to ElastiCache for Redis, especially given the need for fast updates and reads due to the frequent leaderboard changes.
I arrange that the answer is:
Create an Auto Scaling group of c5.large Amazon EC2 On-Demand Instances behind an Application Load Balancer. Use an Amazon ElastiCache for Redis cluster to maintain the leaderboard.