Study guide
Cloud Technology and Services is the largest domain on the exam at roughly a third of scored questions, spanning how customers access AWS, how AWS infrastructure is physically organized, and the breadth of compute, database, networking, storage, and specialty services on offer. This chapter cannot cover every AWS service, but it builds the map of categories and flagship services the exam expects a cloud practitioner to recognize and place correctly.
Accessing AWS and Global Infrastructure
Customers interact with AWS through several methods. The AWS Management Console is a web-based graphical interface for manually configuring and monitoring resources, well suited to learning and one-off tasks. The AWS Command Line Interface (CLI) lets users script and automate actions from a terminal, useful for repeatable operations. Infrastructure as Code (IaC), implemented through services like AWS CloudFormation, defines infrastructure in template files so environments can be created, replicated, and torn down consistently. For connecting a company's existing network to AWS, a Virtual Private Network (VPN) creates an encrypted tunnel over the public internet, suitable for moderate, cost-sensitive connectivity, while AWS Direct Connect establishes a dedicated, private physical network connection between a customer's data center and AWS, offering more consistent bandwidth and lower latency for demanding workloads. AWS organizes its physical infrastructure into Regions, separate geographic areas such as US East (Northern Virginia), each fully isolated from other Regions for fault tolerance and often chosen based on data residency or latency needs. Each Region contains multiple Availability Zones (AZs), which are one or more discrete data centers with independent power, cooling, and networking, connected by low-latency links; running an application across multiple AZs protects it from a single data center failure. Edge locations are smaller sites, part of the Amazon CloudFront content delivery network, that cache content closer to end users to reduce latency. AWS Local Zones extend select Regions with infrastructure placed in additional metropolitan areas for very low-latency needs, and AWS Wavelength embeds AWS compute and storage within telecommunications providers' 5G networks, aimed at applications needing ultra-low latency for mobile end users.
Compute Services
Amazon EC2 (Elastic Compute Cloud) provides resizable virtual servers called instances, available in families optimized for general purpose, compute-intensive, memory-intensive, or storage-intensive workloads, letting a customer pick the balance of CPU, memory, and storage performance their application needs. Containers package an application with its dependencies so it runs consistently across environments; Amazon Elastic Container Service (ECS) is AWS's own container orchestration service, while Amazon Elastic Kubernetes Service (EKS) runs the open-source Kubernetes orchestration platform on AWS for teams standardizing on that ecosystem. AWS Fargate is a serverless compute engine for containers that removes the need to provision or manage the underlying servers for either ECS or EKS, billing based on the resources each task consumes. AWS Lambda runs code in response to triggers, such as a file upload or an API request, without any server management at all, and it only charges for the compute time actually consumed during execution, making it well suited to unpredictable or infrequent workloads. Amazon EC2 Auto Scaling automatically adjusts the number of EC2 instances in a group up or down based on demand or a schedule, helping maintain performance while controlling cost. Elastic Load Balancing (ELB) distributes incoming application traffic across multiple targets, such as EC2 instances in different Availability Zones, improving both fault tolerance and the ability to handle traffic spikes. A scenario describing unpredictable, short-lived workload bursts with no interest in managing servers points toward Lambda; a scenario describing a need for full control over the guest operating system points toward EC2.
Databases and Networking
Amazon RDS (Relational Database Service) manages relational databases such as MySQL, PostgreSQL, or SQL Server, automating patching, backups, and failover. Amazon Aurora is AWS's own MySQL- and PostgreSQL-compatible relational database engine, built for higher throughput and availability than standard RDS engines while remaining compatible with the same tools. Amazon DynamoDB is a fully managed, serverless NoSQL key-value and document database designed for single-digit-millisecond performance at virtually any scale, well suited to session state or shopping cart data. When databases need to move to AWS or convert engines, AWS Database Migration Service (DMS) handles the replication and the Schema Conversion Tool (SCT) converts database schema and code when the source and target engines differ. Networking centers on the Virtual Private Cloud (VPC), a logically isolated section of the AWS cloud where a customer defines its own IP address range, subnets, route tables, and gateways. Amazon Route 53 is a scalable Domain Name System (DNS) web service that translates domain names into IP addresses and can also perform health checks and traffic routing. Amazon CloudFront is AWS's content delivery network, caching content at edge locations worldwide to reduce latency for end users. AWS Global Accelerator improves availability and performance for global applications by routing traffic over the AWS global network backbone rather than the public internet, which helps a scenario describing users scattered across continents connecting to a single application.
Storage Services
Amazon S3 (Simple Storage Service) is object storage with multiple classes trading cost against access frequency and retrieval speed. S3 Standard suits frequently accessed data; S3 Standard-Infrequent Access and S3 One Zone-Infrequent Access cost less for data accessed occasionally, with One Zone trading resilience across Availability Zones for an even lower price; S3 Intelligent-Tiering automatically moves objects between access tiers as usage patterns change, removing the guesswork of picking a class upfront. The Glacier family covers long-term archival: S3 Glacier Instant Retrieval suits archive data that still needs millisecond access, S3 Glacier Flexible Retrieval suits rarely accessed data tolerant of retrieval delays measured in minutes to hours, and S3 Glacier Deep Archive offers the lowest cost for data retained for years to meet compliance requirements, with retrieval taking longer still. Amazon EBS (Elastic Block Store) provides persistent block storage volumes that attach to a single EC2 instance, comparable to a virtual hard drive. Amazon EFS (Elastic File System) provides a fully managed, scalable file system that many EC2 instances can mount and share simultaneously. Amazon FSx offers fully managed file systems built on popular third-party technologies, such as Windows File Server or Lustre, for workloads needing those specific feature sets. AWS Storage Gateway connects on-premises environments to AWS storage, letting existing on-premises applications use cloud storage without a rewrite. AWS Backup centralizes and automates backup policies across many AWS services from a single console rather than configuring backups service by service.
AI/ML, Analytics, and Application Integration Services
AWS offers a range of services for building intelligent and data-driven applications without deep machine learning expertise. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale. Amazon Lex powers conversational interfaces, such as chatbots, using the same natural-language understanding technology behind Amazon Alexa. For analytics, Amazon Athena runs SQL queries directly against data stored in Amazon S3 without needing to load it into a separate database first, ideal for occasional ad hoc analysis of large datasets. Amazon Kinesis ingests and processes streaming data in near real time, such as clickstream data or sensor readings arriving continuously. AWS Glue is a serverless data integration service that discovers, prepares, and combines data for analytics, commonly used to extract, transform, and load (ETL) data between sources. Amazon QuickSight is a business intelligence service for building interactive dashboards and visualizations from underlying data sources. Beyond analytics, several services round out common application-building needs the exam may reference. Amazon SNS (Simple Notification Service) sends messages to many subscribers at once, such as fanning out an alert to email and SMS simultaneously, while Amazon SQS (Simple Queue Service) provides a fully managed message queue that decouples the components of a distributed application so one part failing does not immediately crash another. Amazon Connect provides a cloud-based contact center, and Amazon SES (Simple Email Service) sends and receives bulk or transactional email. Amazon WorkSpaces delivers virtual, cloud-based desktops to end users. AWS Amplify helps developers build and deploy full-stack web and mobile applications quickly. AWS IoT Core connects and manages internet-connected devices at scale, letting them securely send data to and receive commands from AWS.
Key terms
- Availability Zone (AZ)
- — One or more discrete data centers with independent power and networking within an AWS Region, used to build fault-tolerant applications.
- AWS Direct Connect
- — A dedicated, private physical network connection between a customer's location and AWS, offering more consistent performance than a VPN.
- AWS Local Zones
- — Infrastructure deployments that extend an AWS Region into additional metropolitan areas for lower-latency access.
- Amazon EC2
- — A service providing resizable virtual server instances in various families optimized for different workload needs.
- AWS Lambda
- — A serverless compute service that runs code in response to triggers and charges only for compute time actually consumed.
- Amazon DynamoDB
- — A fully managed, serverless NoSQL key-value and document database built for high-scale, low-latency access.
- Amazon VPC
- — A logically isolated virtual network within AWS where a customer controls IP ranges, subnets, and routing.
- Amazon CloudFront
- — AWS's content delivery network that caches content at edge locations to reduce latency for end users.
- Amazon S3
- — Object storage offering multiple classes that trade cost against access frequency, from Standard to the Glacier archival tiers.
- Amazon Athena
- — A serverless service for running SQL queries directly against data stored in Amazon S3.
- Amazon SNS
- — A fully managed publish/subscribe messaging service that delivers messages to many subscribers at once.
- Amazon SQS
- — A fully managed message queuing service that decouples components of a distributed application.
Exam tips
- Distinguish VPN (encrypted tunnel over the public internet) from Direct Connect (dedicated private physical connection) by the level of consistency and bandwidth the scenario needs.
- Remember the infrastructure hierarchy: Regions contain Availability Zones, Availability Zones contain data centers, and edge locations sit outside this hierarchy to serve CloudFront content.
- Match compute services to the amount of server management implied: EC2 (full control), ECS/EKS (container orchestration), Fargate (serverless containers), Lambda (serverless functions, no servers at all).
- S3 storage class questions almost always hinge on how frequently data is accessed and how fast it must be retrievable — map the scenario's access pattern to the cheapest class that still meets it.
- SNS pushes messages out to many subscribers (fan-out); SQS holds messages in a queue for one or more consumers to pull and process — a frequent point of confusion on the exam.