A company is developing a solution to analyze images. The solution uses a 50 TB reference dataset and analyzes images up to 1 TB in size. The solution spreads requests across an Auto Scaling group of Amazon EC2 Linux instances in a VPC. The EC2 instances are attached to shared Amazon EBS io2 volumes in each Availability Zone. The EBS volumes store the reference dataset. During testing, multiple parallel analyses led to numerous disk errors, which caused job failures. The company wants the solution to provide seamless data reading for all instances. Which solution will meet these requirements MOST cost-effectively?