Which solution will meet these requirements with the LEAST latency?
Create a two-node DynamoDB Accelerator (DAX) cluster. Configure an application to read and write data by using DAX.
Create a three-node DynamoDB Accelerator (DAX) cluster. Configure an application to read data by using DAX and to write data directly to the DynamoDB table.
Create a three-node DynamoDB Accelerator (DAX) cluster. Configure an application to read data directly from the DynamoDB table and to write data by using DAX.
Create a single-node DynamoDB Accelerator (DAX) cluster. Configure an application to read data by using DAX and to write data directly to the DynamoDB table.
Explanations:
A two-node DAX cluster does not provide high availability due to the lack of redundancy. It is also not sufficient to support the high throughput needs of a latency-sensitive trading platform.
A three-node DAX cluster ensures high availability and can handle increased read requests efficiently. Using DAX for reads improves performance while writing directly to DynamoDB allows for immediate data consistency.
While a three-node DAX cluster provides high availability, configuring it to read directly from DynamoDB and write through DAX adds unnecessary latency for reads, which defeats the purpose of optimizing for performance.
A single-node DAX cluster does not offer high availability, making it a risky choice for a trading platform that requires consistent uptime. Using DAX only for reads limits performance optimization.
I appraise that the answer is:
Create a three-node DynamoDB Accelerator (DAX) cluster. Configure an application to read data by using DAX and to write data directly to the DynamoDB table.