AWS Service Study Guides

Durable object storage with tiered pricing

Amazon S3 is object storage built for 99.999999999% (11 nines) durability. Objects live in buckets within a Region and are replicated across multiple Availability Zones. Storage classes (Standard, Intelligent-Tiering, Standard-IA, One Zone-IA, Glacier tiers, Deep Archive) trade cost against access frequency and retrieval time.

Block storage volumes for EC2

Amazon EBS provides persistent block-level volumes attached to EC2 instances in a single AZ. Families: gp3 (general SSD), io2/io2 Block Express (provisioned IOPS), st1 (throughput HDD), sc1 (cold HDD). Snapshots stored in S3 enable backup and cross-AZ/Region copies.

Shared, elastic NFS file storage

Amazon EFS is a managed NFS file system that scales elastically and is accessible from multiple AZs. Mount targets in each AZ let many instances share the same file system concurrently.

Managed relational databases

Amazon RDS manages relational engines (PostgreSQL, MySQL, MariaDB, Oracle, SQL Server) — handling provisioning, patching, backups, and failover. Multi-AZ gives synchronous standby failover; read replicas scale reads asynchronously.

Cloud-native relational engine (MySQL/PostgreSQL compatible)

Amazon Aurora is a MySQL- and PostgreSQL-compatible relational database built for the cloud, with storage that auto-scales across three AZs and up to 15 read replicas. Aurora Serverless v2 scales capacity automatically for variable loads.

Serverless NoSQL at any scale

Amazon DynamoDB is a fully managed key-value and document database delivering single-digit-millisecond performance at virtually any scale. Capacity modes: on-demand (per-request billing) and provisioned (with auto scaling). DAX adds microsecond in-memory caching; global tables replicate across Regions.

Managed in-memory caching

Amazon ElastiCache provides managed Redis OSS and Memcached in-memory data stores. Used for sub-millisecond caching, session stores, leaderboards, and offloading repeated database reads via cache-aside patterns.

Event-driven serverless functions

AWS Lambda runs code in response to events with no servers to manage. Billing is per request plus duration (GB-seconds). Functions scale automatically to thousands of concurrent executions and integrate natively with S3, DynamoDB, SQS, API Gateway, EventBridge, and more.

Virtual servers with full control

Amazon EC2 provides resizable virtual machines with complete OS control. Pricing models include On-Demand, Savings Plans, Reserved Instances, Spot (up to ~90% off, interruptible), and Dedicated Hosts (for licensing/compliance). Placement groups and instance types tune performance.

Managed message queues for decoupling

Amazon SQS is a fully managed message queue that decouples producers and consumers. Standard queues maximize throughput with at-least-once delivery; FIFO queues preserve order with exactly-once processing. Dead-letter queues isolate failing messages.

Pub/sub messaging and fan-out

Amazon SNS is a managed publish/subscribe service that fans messages out to many subscribers (SQS queues, Lambda, HTTP endpoints, email, SMS). The SNS→SQS fan-out pattern gives each consumer its own durable queue.

Global CDN with edge caching

Amazon CloudFront is a content delivery network that caches content at global edge locations, reducing latency and offloading origins. It integrates with S3, ALBs, API Gateway, and supports HTTPS via ACM, geo-restriction, signed URLs, and WAF.