A company hosts a photography website on AWS that has global visitors. The website has experienced steady increases in traffic during the last 12 months, and users have reported a delay in…

Setting the Minimum TTL and Maximum TTL to 0 would prevent CloudFront from caching content at the edge, forcing every request to go back to the origin, which would significantly increase latency rather than reduce it.

Setting a high Minimum TTL and Maximum TTL in the CloudFront distribution allows images to be cached at CloudFront edge locations for a longer duration, maximizing cache hits and significantly reducing latency for subsequent requests as content is served directly from the nearest edge, rather than the origin.

Setting the CloudFront distribution to forward all headers, all cookies, and all query strings would reduce the cache hit ratio because CloudFront would create more unique cache keys or bypass caching, leading to more requests hitting the origin and increasing latency.

Setting up additional origin servers that are geographically closer to different groups of requesters ensures that even if a cache miss occurs, or for content that is not cached, the request travels a shorter distance to the nearest origin server, thereby reducing the overall end-to-end latency for content delivery to global visitors.

Selecting Price Class 100 on the CloudFront distribution extends content delivery to all CloudFront edge locations, including the most remote, but it's primarily a cost-versus-reach configuration and does not directly optimize caching efficiency or origin response times for minimal latency as effectively as high TTLs and geographically distributed origins.