- 2073 Bhadra · 8+2 marks
CDN reducing delay in object retrieval
Short answer
A CDN caches content on edge servers near users, cutting propagation delay and origin load via DNS redirection. Not all objects benefit: dynamic, uncacheable, or rarely requested content still requires origin trips, and cold misses add an extra hop.
Describe how content distribution network reduces the delay in receiving a requested object. Will content distribution reduce the delay for all objects requested by a user? Explain your answer with appropriate figure.
Role of a CDN
A Content Distribution Network (CDN), such as Akamai or Cloudflare, is a set of geographically distributed servers (edge or surrogate servers) that store copies of web content. Instead of every user fetching an object from the distant origin server, the request is directed to a nearby CDN server.
How a CDN reduces delay
- Short distance. The content is physically close to the user, so propagation delay and the number of hops are reduced.
- Avoids bottlenecks. Traffic does not cross congested links to the origin server, and the user's path to the nearest edge server is short and lightly loaded.
- Load sharing. Many servers share the requests, so no single server is overloaded and queuing delay is low.
- Caching. Popular objects stay in the edge cache; the origin is contacted only on a miss or expiry.
- Smart redirection. DNS (or an HTTP redirect) returns the IP address of the best server, chosen by network proximity or load.
Origin server (far away)
| (only on cache miss)
v
+-------------+ +-------------+
| CDN edge A | | CDN edge B |
+------^------+ +------^------+
| |
User 1 User 2
Operation: the user's browser asks for www.site.com/video. The site's DNS (a CNAME) points to the CDN's DNS. The CDN DNS picks the nearest server and returns its IP address. The browser fetches the object from that edge server. On a miss, the edge server fetches it from the origin, stores it and serves later users from its cache.
Does a CDN reduce delay for all objects?
No.
- Dynamic or personalised content (bank balance, search results, real-time data) cannot be cached, so each request still goes to the origin server.
- Rarely requested (unpopular) objects are usually not in the edge cache; the first request is a miss and takes even longer because of the extra hop through the edge server.
- If the user is already near the origin, the gain is small.
- Objects that have expired or are marked non-cacheable must be revalidated with the origin.
So a CDN helps mainly for static, popular and large objects such as images, video, scripts and style sheets.
✓ Answer checked by Claude Opus 5.5 (high effort)
Related questions
- Redundant network design for insurance company · Internet and Intranet, chapter 5
- Redundant network for IOE Pulchowk campus · Internet and Intranet, chapter 5
- Reducing power loss in distribution systems · Transmission and Distribution System Design, chapter 7 · asked 2×
- Information retrieval in cloud · Information System, chapter 7
- Radial, loop and network distribution merits · Transmission and Distribution System Design, chapter 5
- Flow distribution in five-pipe network · Hydraulics, chapter 3