Google Cloud CDN Cheat Sheet

Google offers two content delivery network services:

• Cloud CDN: Accelerates web applications and APIs using Google’s global edge network.
• Media CDN: Built on YouTube’s infrastructure for video streaming (VoD and live) and large file downloads.

Both services help reduce latency, cost, and load on backend services.

Features

• Activates with a single click for Cloud Load Balancing users.
• Cloud CDN supports modern protocols originally developed at Google, like HTTP/2 and QUIC.
• Integrates with Cloud Monitoring and Cloud Logging by providing latency metrics and raw HTTP request logs for deeper and better visibility.
• Logs can be exported to Cloud Storage or BigQuery for analysis.
• Cloud CDN content can be sourced from several types of backends including:
  • Instance groups
  • Zonal network endpoint groups (NEGs)
  • Serverless NEGs: One or more App Engine, Cloud Run, or Cloud Functions services
  • Internet NEGs, for endpoints that are outside of Google Cloud (also known as custom origins)
  • Buckets in Cloud Storage

• Cloud CDN also delivers content hosted on-premises or in another cloud over Google’s high-performance distributed edge caching infrastructure.
• Fine-grain cache controls: Configure caching behavior by origin with control over cache keys, TTLs, and other settings based on content type.
• Advanced HTTP routing: Map traffic to specific edge configurations and origins at a fine-grained level.
• Modern protocols: Support for HTTP/2, QUIC, and TLS 1.3 for improved performance.
• Integration with Cloud Armor: Protect content at the edge with DDoS mitigation and security policies.
• Integration with Service Extensions: Add custom code to the request processing path using plugins.
• Cache hits and misses: Requests served from cache (hits) reduce latency; requests forwarded to origin (misses) fill the cache.
• Cache fill: Data transfer from origin to cache incurs charges based on geographic regions.
• Cache invalidation: Remove cached content manually when needed.
• Real-time monitoring: View cache hit ratios, latency metrics, and request logs via Cloud Monitoring and Cloud Logging.

How It Works

• Cloud CDN works with the global external Application Load Balancer or classic Application Load Balancer.
• Requests arrive at a Google Front End (GFE) at the edge of Google’s network closest to the user.
• Cache hit: Content served from cache – faster response, lower cost, reduced origin load.
• Cache miss: Request forwarded to origin server – response stored in cache for future requests (cache fill).
• Partial hit: Part of the request served from cache, part from origin (for byte range requests).
• Cache hit ratio: Percentage of requests served from cache. Monitor in Cloud Console over periods from minutes to 30 days.
• Eviction: Cached content removed to make space for newer content (based on popularity, not expiration time).
• Expiration: Content removed after TTL (time-to-live) even if still cached.
• You cannot preload caches; cache fill happens only in response to client requests.

Pricing

Cloud CDN pricing is based on bandwidth usage and HTTP/HTTPS requests.

• Cache egress (outbound data transfer): Charged per GiB, with tiered rates based on monthly usage volume.
• Cache fill (data transfer from origin to cache): Charged per GiB, with rates varying by geographic region.
• Lookup requests: Charged per 10,000 HTTP/HTTPS cache lookup requests.
• Cache invalidations: You are charged for manually initiated cache invalidations.

For current pricing details and volume discounts, refer to the official Google Cloud CDN pricing page.

Google Cloud CDN Cheat Sheet References:

https://cloud.google.com/cdn/
https://cloud.google.com/cdn/docs/overview