AWS: CSA Flashcards
List CloudWatch EC2 metrics that are available by default
- CPU utilisation
- Network utilisation
- Disk reads
List CloudWatch EC2 metrics that are not readily available by default
- Memory utilisation
- Disk swap utilisation
- Disk space utilisation
- Page file utilisation
- Log collection
What is Amazon FSx for Lustre?
- High-performance file system built on Lustre
- Lustre is an open-source parallel file system
- Stores data across multiple network file servers to maximise performance and reduce bottlenecks
- Use cases
- High performance computing where high throughput and low latency is essential for processing large datasets
- Machine learning and analytics
- Media processing workloads (eg. video rendering, transcoding) where fast access to large files is required
What is Amazon FSx for Windows FS?
- Fully managed, high performance file storage service compatible with Windows
- Supports SMB protocol, Windows NTFS and Microsoft Active Directory integration
- Thousands of compute instances can access a file system concurrently
- Use cases:
- Enterprise applications eg. Microsoft SharePoint, Exchange and Active Directory
- Migrating existing Window-based applications to AWS
- Line of business applications
What are the options to provide user authentication and access control for your file system if using Amazon FSx for Windows?
- Options:
- AWS managed Microsoft Active Directory
- Self managed Microsoft Active Directory
- After creating an AD config for a file system, the config can’t be changed
- You will need to create a new file system from a backup and change the AD config for that file system
- The configs allow users in your domain to use their existing identity to access FSx file system and control access to individual files and folders
What is Amazon EFS?
- Fully managed file storage service for Linux-based applications
- Supports NFSv4 - easy to mount EFS on multiple EC2 instances simultaneously
- High availability and durability
- Data stored across multiple AZ within a region
- Built-in data redundancy and automatic failover
- Use cases:
- Containerised applications - shared storage for containers running on multiple EC2 instances
- CMS - multiple web servers can share access to same files and data
- Dev and test environments - devs can share code/resources across multiple instances
What happens when EFS is mounted on EC2 instances?
- Provides file system interface
- Multiple EC2 instances can access an EFS file system at the same time, allowing EFS to provide a common data source for workloads and applications running on more than one EC2 instance
What is AWS Glue?
- Fully managed ETL service
- Serverless environment for running ETL jobs
- Pay for resources only during job execution
- Provides a visual interface for defining ETL jobs
- Data can be extracted from various sources and transformed to a suitable format for analysis
- Automatic schema discovery and mapping
- Use cases:
- Data integration and transformation
- Data lakes and data warehouses
- Serverless data processing
What is AWS DMS?
- Fully managed database migration service
- Supports homogeneous and heterogeneous migrations
- Supports schema and data transformations
- Map data to different schema structures
- Transform data as it is being migrated
- Continuous data replication
- Perform one-time migrations as well as ongoing replication to keep databases in sync
- Use cases:
- Cloud migration
- Database consolidation
- Disaster recovery
What is AWS SCT?
- Standalone schema conversion tool
- Automate process of converting db schemas from one db engine to another
- Analyzes source schema and generates target schema compatible with chosen target db engine
- Use cases:
- Database migration
- Database engine upgrades
- Cross-platform database consolidation
What is AW Elastic Beanstalk?
- Fully managed platform-as-a-service
- Developer-centric view of deploying an app on AWS
- Once deployed, it builds the selected supported platform version and provisions one/more AWS resources (ie. EC2 instances) to run the application
- Free service but underlying AWS resources will have costs involved
- Use cases:
- Web applications
- Microservices
- DevOps workflows
What is Lambda@Edge?
- Serverless compute service extending Lambda capabilities to CloudFront edge locations
- Execute code closer to your end users to reduce latency
- Leverages Lambda service to automatically scale functions
- Event-driven - functions run in response to CloudFront events
- Use cases:
- Dynamic content personalisation
- Security and access control
- Protect against XSS and SQL injections
- Content optimisation
- Optimise content delivery by compressing images, caching frequently accessed resources
At which points can you use Lambda functions to change CloudFront requests/responses?
- Viewer request - after CloudFront receives a request from a viewer
- Origin request - before CloudFront forwards request to origin
- Origin response - after CloudFront receives response from origin
- Viewer response - before CloudFront forwards response to viewer
How does Lambda work with CloudFront CDN?
- CloudFront functions
- Lambda@Edge
What are the differences between CloudFront functions vs Lambda@Edge?
- CloudFront functions
- Written in JS
- Limited integration with AWS services - focused on CDN-related tasks
- Runtime limitations - designed for lightweight, short-lived tasks that are executed quickly at the edge
- Can be triggered by viewer requests/responses
- Lambda@Edge
- Supports multiple programming languages
- Can be integrated with a lot of AWS services
- Fewer runtime limitations - suitable for more complex and resource-intensive tasks
- Can be triggered by all requests/responses from CloudFront (incl. origin requests/responses)
What is the difference between signed URLs vs signed cookies when serving private content from CloudFront?
- Signed URL:
- Embed authentication information directly into URLs making it useful for one-time or temp access to specific resources
- Use cases:
- RTMP distribution
- Signed cookie:
- Store authentication information in user’s browser to maintain session-based authentication and authorisation
- Use cases:
- Provide access to multiple restricted files
- Don’t want to change current URLs
What is AWS Shield?
- Managed DDoS protection service
- Automatically included at no extra cost with all AWS services that are accessible over the internet ie. CloudFront, ELB, Route53
- Third-party DDoS providers may offer more customisation options but require more setup and may incur more costs
- Use cases:
- Protecting web apps
- Ensuring high availability
- Safeguarding against financial loss (Shield Advanced)
What is the difference between AWS Shield - standard vs advanced?
- Standard
- Automatic protection against most common DDoS attacks
- Included with all AWS services at no extra cost
- Advanced
- All features of standard
- Real-time attack visibility and metrics
- Additional cost protection features to safeguard against financial loss
- Personalised support during attacks
What is AWS Lake Formation?
- Simplifies setting up and managing data lakes on AWS
- Automates data ingestion, transformation, and access controls
- Supports batch and real-time data ingestion
- Includes a centralised data catalog that automatically indexes and organises metadata so it’s easy to search and analyze
- Use cases:
- Analytics and machine learning where there are large volumes of data from various sources
- ETL jobs
- Data governance and compliance
What are data lakes?
Centralised repo that allows you to store all your structured/unstructured data at any scale
How does AWS Lake Formation define access control policies?
- IAM policies
- Resource-based policies
- Used to grant access to specific AWS accounts, IAM roles or federated users
- Granular permissions
- Granular permissions can be granted at the db, table, column level
- Using the “grant” api operation
- When granting permissions, you specify the resource (eg. tables, columns) and the actions the user can perform
What is the difference between FSx for Lustre vs EFS?
- FSx for Lustre
- Designed for high performance, compute-intensive workloads
- EFS
- General purpose file system suitable for a range of use cases
What is the difference between EFS vs S3?
- EFS
- Shared file storage mounted directly to EC2 instances
- Suitable for applications/workloads that require shared access to files
- S3
- Object storage service
- Store and retrieve large amounts of unstructured data ie. images/videos/backups
What is AWS RAM?
- Service that enables you to securely share your AWS resources with other AWS accounts
- Accounts within your organisation or outside
- Avoid overhead and complexity of duplicating resources across multiple accounts and regions
- Centralised management and monitoring
- Use cases:
- Shared network resources
- Shared DNS rules
- Cost efficiency
What is AWS Organisations?
- Management service that enables you to consolidate AWS accounts
- Provides framework for applying policies and permissions across accounts to simplify billing and enhance security
- Use cases:
- Consolidating billing
- Centralised policy management
- Account isolation - separate accounts to reduce risk of affecting production accidentally
What is the difference between AWS Organisations vs AWS Control Tower?
- Organisations
- Granular control and customisation options for advanced users
- Suitable for users wanting detailed control over their account structures and policies
- Control tower
- Uses AWS Organisations under the hood
- User-friendly with guided setup
- Suitable for users wanting streamlined setup
What is AWS Secret Manager?
- Service for securely storing, managing and retrieving secrets
- Secrets are stored securely using encryption keys managed by AWS KMS
- Secrets are encrypted at rest and in transit
- Automated secret rotation
- Supports multiple versions of a secret
- Use cases:
- Storing db credentials, config secrets and API keys
- Handle secrets in microservices architecture by enabling each service to only access the secrets it needs