3.1 - High-performing content delivery (CloudFront)
AWS Solutions Architect Associate objective 3.1 covers designing for high-performing content delivery and networking. Amazon CloudFront is the content delivery network (CDN) that caches static and dynamic content at edge locations close to users, reducing latency and offloading the origin - a global website serving mostly static assets gets lower worldwide latency this way. When a CloudFront site must run custom logic such as A/B routing, header manipulation or authorization at the edge, Lambda@Edge functions run that code at edge locations. You should also know Global Accelerator for TCP/UDP performance, edge caching behaviours, and origin choices like S3 and load balancers. Expect scenario questions that describe reducing global latency for cached content or running custom logic close to users, and ask which service - CloudFront or Lambda@Edge - and which design achieves the performance goal.
Cache content at edge, lower global latency = CloudFront (CDN). Run custom logic (A/B, auth) at the edge = Lambda@Edge. TCP/UDP performance over the AWS backbone = Global Accelerator.
Practice questions
1. Which service caches static and dynamic content at edge locations to reduce global latency for a website?
- Amazon Route 53
- Amazon CloudFront (correct answer)
- AWS Global Accelerator
- Elastic Load Balancing
CloudFront is a CDN that caches content at edge locations near users, cutting latency and offloading the origin. Global Accelerator improves network path but does not cache content.
2. Which DynamoDB feature returns cached reads in microseconds without changing application query code?
- Global secondary index
- On-demand capacity mode
- DynamoDB Accelerator (DAX) (correct answer)
- Time to Live (TTL)
DAX is an in-memory cache in front of DynamoDB delivering microsecond reads with an API-compatible client, so app code barely changes. GSIs add query patterns; TTL expires items.
3. Which EBS volume type is the best general-purpose default balancing price and performance for most workloads?
- gp3 (General Purpose SSD) (correct answer)
- io2 Block Express (Provisioned IOPS)
- sc1 (Cold HDD)
- st1 (Throughput Optimized HDD)
gp3 is the default general-purpose SSD, offering a strong price/performance balance with independently configurable IOPS and throughput. io2 targets high-IOPS critical databases; HDD types suit sequential workloads.
4. A relational database is overwhelmed by read traffic while writes stay light. Which change scales reads with the least application rework?
- Switch the database to DynamoDB
- Add Amazon RDS read replicas (correct answer)
- Enable Multi-AZ on the primary
- Move the database to an instance store
Read replicas offload read queries to one or more asynchronous copies, scaling reads without redesigning the app. Multi-AZ is for availability, not read scaling; DynamoDB is a larger migration.
5. An unpredictable-traffic API on DynamoDB must never be throttled and should not require capacity planning. Which capacity mode fits?
- On-demand capacity mode (correct answer)
- Provisioned capacity with fixed units
- Provisioned with no auto scaling
- Reserved capacity purchased upfront
On-demand mode scales instantly to traffic and bills per request, ideal for spiky or unknown workloads with no capacity planning. Fixed provisioned units throttle when exceeded; reserved capacity assumes steady, predictable usage.
6. An application needs host-based and path-based HTTP routing with support for redirects and WebSocket. Which load balancer should be used?
- A Network Load Balancer operating at Layer 4 for raw TCP throughput
- A Gateway Load Balancer forwarding traffic to virtual appliances
- A Classic Load Balancer kept for its legacy Layer 7 features
- An Application Load Balancer routing on host and path at Layer 7 (correct answer)
The ALB is the Layer 7 load balancer supporting host/path-based routing, HTTP redirects, header/query rules, and WebSocket/HTTP2. NLB is Layer 4; GWLB is for inline appliances.