Which component in a CDN is primarily responsible for storing cached content close to users to reduce latency?
Think about which part of the CDN is physically near the user.
Edge servers cache content near users to reduce latency and improve load times. Origin servers hold the original content, load balancers distribute traffic, and DNS resolvers translate domain names.
When a CDN experiences a sudden global spike in traffic, which strategy best helps maintain performance?
Think about distributing load closer to users worldwide.
Deploying more edge servers globally distributes traffic and reduces latency. Adding origin servers in one region doesn't help global users, increasing cache time may cause stale content, and reducing edge servers worsens performance.
What is the main tradeoff when setting a very short cache expiration time on CDN edge servers?
Consider how often edge servers must fetch fresh content.
Short cache expiration means edge servers refresh content often, reducing latency for fresh content but increasing load on origin servers. Longer expiration reduces origin load but risks stale content.
In a typical CDN request flow, what happens if the requested content is not found in the edge server cache?
Think about how the CDN obtains content not locally cached.
If the edge server cache misses, it fetches the content from the origin server, caches it, and then serves it to the client. It does not return an error or rely on DNS or client retries.
A CDN edge server handles 10,000 requests per second, each averaging 500 KB. Estimate the minimum network bandwidth in Gbps required to support this load without congestion.
Calculate total data per second and convert bytes to bits.
10,000 requests/sec * 500 KB = 5,000,000 KB/sec = 5 GB/sec. 5 GB/sec * 8 (bits/byte) = 40 Gbps minimum bandwidth.