Is Using a CDN Always Faster? A Deep Dive into How CDN Works and When It Helps

What Is a CDN?

For textual data developers typically store information in a MySQL database and add a Redis cache layer to reach read performance of tens of thousands QPS. For binary objects such as images or videos, the storage layer switches to an object storage service (e.g., Amazon S3 or Alibaba Cloud OSS) and the cache layer becomes a Content Delivery Network (CDN). The CDN acts as the cache for object storage.

How CDN Works

A CDN‑served image URL looks like https://cdn.xiaobaidebug.top/1667106197000.png. The domain part ( cdn.xiaobaidebug.top) resolves to a CDN node IP through a multi‑step DNS process:

Step 1‑2: Browser and OS cache are checked; if missing, the local DNS resolver (e.g., home router) is queried.

Step 3: If the resolver has no answer, it queries the root, TLD, and authoritative DNS servers.

Step 4: The authoritative server returns a CNAME record such as cdn.xiaobaidebug.top.w.kunlunaq.com, pointing to the CDN provider’s DNS scheduler.

Step 5‑7: The scheduler returns the IP address of the CDN node that is "nearest" to the client.

The "nearest" node is chosen by the CDN‑specific DNS server, which knows the geographic location, network latency, and load of each CDN edge server.

Example DNS queries:

$ dig +trace xiaobaidebug.top
xiaobaidebug.top. 600 IN A 47.102.221.141

$ dig +trace cdn.xiaobaidebug.top
cdn.xiaobaidebug.top. 600 IN CNAME cdn.xiaobaidebug.top.w.kunlunaq.com.
cdn.xiaobaidebug.top.w.kunlunaq.com. 300 IN A 122.228.7.243
...

Back‑origin (回源)

When a CDN node does not have the requested file (cache miss) it fetches the object from the origin storage (OSS) and stores it locally. The HTTP response header X-Cache indicates whether the request hit the cache: MISS TCP_MISS – cache miss, CDN fetched from origin. HIT TCP_MEM_HIT – cache hit, data served directly from the CDN node.

Using tools such as Postman or curl to inspect the X-Cache header lets you calculate the back‑origin ratio (e.g., 90% miss means the CDN is rarely helping).

Is Using a CDN Always Faster?

If the request hits the CDN cache, latency is reduced because the data is served from a node close to the user. However, a cache miss adds an extra round‑trip to the origin storage, making the request slower than a direct fetch. Therefore, CDN acceleration only helps when the cache‑hit rate is high.

When a CDN is introduced but the cache is empty or expired, the flow becomes: client → DNS → CDN node → origin → CDN node → client, which is slower than a direct client → origin request.

When Not to Use a CDN

Intranet services where the client, application server, and object storage reside in the same data center; the "nearest" CDN node offers no latency benefit.

Files that are rarely accessed or not repeatedly requested; each request will likely miss the CDN cache and incur a back‑origin fetch, adding unnecessary overhead.

In such cases, measuring the X-Cache miss ratio can justify removing the CDN layer.

Optimization Tips

Pre‑warm hot objects by proactively requesting them so they are cached on edge nodes before traffic spikes.

Use gradual rollout (gray release) to seed the CDN cache with real user traffic.

Increase cache TTL for frequently accessed hot data to reduce back‑origin frequency.

Conclusion

Textual data: MySQL + Redis; binary objects: Object Storage + CDN.

CDN accelerates delivery only when the cache is hit; a miss results in slower response due to back‑origin.

The CDN’s main advantage is geographic proximity and cache reuse for repeated accesses. If your service is internal or the files are seldom reused, CDN adds cost without benefit.