DJames617 A service that creates, stores, and renews TLS certificates for supported AWS services. TLS certificates enable encrypted HTTPS connections.
AWS Study Notes / Beginner Reference
AWS Glossary
Plain-language definitions for the AWS services, acronyms, and architecture terms that appear throughout the Study Notes and AWS Arcade.
Start here if AWS names and abbreviations are unfamiliar. Learn the purpose of a term first, then return to the certification guide to practice choosing it under scenario constraints.
A Layer 7 load balancer for HTTP and HTTPS traffic. It can route by host name, path, headers, or query strings and is commonly used with containers and web applications.
A managed front door for APIs. It can provide routing, authentication or authorization integrations, throttling, stages, monitoring, and request handling without operating a traditional web server.
An isolated location inside an AWS Region. Deploying across multiple AZs helps an application survive the loss of one data-center area.
AWS infrastructure as code. A template describes resources and their configuration so environments can be deployed repeatedly and reviewed as changes.
AWS content delivery network (CDN). It caches and serves content from edge locations closer to users, while an origin such as S3, ALB, or API Gateway provides the source content.
An audit service that records AWS API activity, including who or what made a call, which API was called, and when. It is not the same as application metrics or logs.
AWS monitoring for metrics, logs, alarms, dashboards, and operational signals. CloudWatch answers questions such as what is happening now or whether a threshold was crossed.
A managed NoSQL database optimized for key-value and document access at scale. Good designs begin with access patterns and partition keys rather than relational joins.
Virtual machines in AWS. EC2 provides substantial operating-system and software control, but the customer manages more patching, scaling, networking, and capacity decisions.
Persistent block storage volumes attached to EC2. EBS volumes are generally Availability-Zone scoped, so snapshots and replication choices matter for recovery and resilience.
AWS container orchestration. ECS schedules and manages containers; with Fargate, AWS provides the underlying container compute so the customer does not manage EC2 worker hosts.
Managed elastic file storage using the NFS protocol. It is designed for shared file access from multiple compute resources and can span Availability Zones within a Region.
A serverless compute engine for containers used with ECS or EKS. You manage images, task definitions, roles, networking, and application security, but not the container host operating systems.
The AWS service for identities and permissions. IAM users, roles, policies, and trust relationships determine who or what can perform actions on resources.
An assumable identity that provides temporary credentials. Roles are preferred for AWS workloads, federation, and cross-account access because applications do not need long-lived access keys.
A managed service for creating and controlling encryption keys. IAM permissions and KMS key policies both matter when deciding who can use a key.
Serverless function compute. You deploy code that runs in response to events and pay mainly for requests and execution duration, while accepting runtime, duration, and concurrency limits.
A stateless, subnet-level network filter. NACLs support allow and deny rules, and return traffic must be explicitly allowed.
A Layer 4 load balancer for TCP, UDP, and TLS workloads. It is used when transport-level handling, static IPs, or very high performance is more important than HTTP-aware routing.
A managed service for common relational database engines. AWS manages much of the infrastructure, while the customer still chooses engine settings, access controls, backups, scaling, and data design.
A geographic AWS location containing multiple isolated Availability Zones. Region choice affects latency, data residency, service availability, disaster recovery, and transfer cost.
The maximum acceptable amount of data loss measured in time. A one-hour RPO means the recovery design must limit data loss to about one hour or less.
The maximum acceptable time to restore service after an interruption. RTO is about downtime, while RPO is about data loss.
Durable object storage for files, logs, static assets, backups, and data lakes. Access is controlled through IAM, bucket policies, encryption, and related controls.
A managed queue for buffering work, absorbing bursts, retrying messages, and decoupling producers from consumers. Design for at-least-once delivery unless using FIFO features and still make consumers idempotent.
A managed pub/sub service for publishing a message to multiple subscribers. SNS is commonly the fanout choice; SQS is commonly the durable work-queue choice.
A logically isolated network in AWS. Subnets, route tables, security groups, NACLs, gateways, and endpoints determine how resources communicate.
A filter for HTTP and HTTPS requests. WAF rules can inspect requests and block or allow patterns such as malicious input, unwanted IPs, or rate-based abuse.
AWS secures the infrastructure and managed-service boundary; the customer remains responsible for items such as identities, data, configuration, permissions, guest operating systems, and application code according to the service used.
An availability design that places redundant resources or managed database capacity in multiple Availability Zones. It primarily addresses AZ failure, not every disaster-recovery requirement.
A replicated database copy used primarily for read scaling, reporting, or sometimes disaster recovery. It is not automatically the same as a synchronous Multi-AZ failover standby.
Grant only the actions, resources, and conditions a principal needs. Least privilege is an access-design principle, not a separate AWS service.
A property in which repeating the same request does not create an unintended additional effect. It is essential when queues, retries, or distributed systems may deliver work more than once.