System Design: The Content Delivery Network (CDN)

Serving distant users from one origin can add network delay and concentrate delivery load. This chapter designs a content delivery network (CDN) to place reusable content nearer its readers, then examines the routing, freshness, and failure requirements that remain.

Problem statement

Consider a data-intensive application serving millions of global users from a single data center. This centralized architecture faces several critical challenges:

• Latency: Separate transmission time, signal propagation, queuing, and processing when tracing a request. Interactive voice and buffered video have different timing needs. For a hypothetical interactive application, a 150-ms target must specify whether it measures one-way delivery, a round trip, or an end-to-end user action; do not apply it to all streaming workloads.

Note: According to readings from December 2021, the average round-trip latency from US East (N. Virginia) to US West (N. California) was 62.9 ms. Intercontinental latency, for example, from US East to Africa (Cape Town), was 225.63 ms.